Single Technology Appraisal

Baricitinib for treating severe alopecia areata [ID3979]

Committee Papers

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

SINGLE TECHNOLOGY APPRAISAL

Baricitinib for treating severe alopecia areata [ID3979]

Contents:

The following documents are made available to stakeholders:

The final scope and final stakeholder list are available on the NICE website.

1. Company submission summary from Eli Lilly

2. Clarification questions and company responses

3. Patient group, professional group and NHS organisation submissions from:

• a. Alopecia UK

• b. British Association of Dermatology

4. External Assessment Report prepared by BMJ-TAG

5. External Assessment Report – factual accuracy check

6. Technical engagement response from company

7. Technical engagement responses and statements from experts:

  - **a.** Dr Abby MacBeth, consultant dermatologist – clinical expert, nominated by British Association of Dermatology 

  - **b.** Dr Matthew Harries, consultant dermatologist – clinical expert nominated by British Association of Dermatology 

  - **c.** Lynn Wilks, trustee and volunteer – patient expert, nominated by Alopecia UK 

  - **d.** Sue Schilling, chief executive officer – patient expert, nominated Alopecia UK 

8. Technical engagement responses from stakeholders:

• a. British Association of Dermatology

9. External Assessment Report critique of company response to technical engagement prepared by BMJ-TAG

• Any information supplied to NICE which has been marked as confidential, has been redacted. All personal information has also been redacted.

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

Single technology appraisal

Baricitinib for treating severe alopecia areata [ID3979]

Document B

Company evidence submission

August 2022

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Contents

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Tables and figures

Table 1 The decision problem ....................................................................................................... 11 Table 2: Technology being evaluated ............................................................................................ 13 Table 3. Types of alopecia areata ................................................................................................. 16 Table 4. Comorbid diseases commonly reported in patients with AA. .......................................... 22 Table 5. Clinical effectiveness evidence ........................................................................................ 29 Table 6: Summary of methodology for BRAVE-AA1 and BRAVE-AA2 ........................................ 35 Table 7. Definitions of outcomes used in BRAVE-AA1 and BRAVE-AA2 ..................................... 38 Table 8. Interpretation of the SALT score ...................................................................................... 40 Table 9. Baseline and disease characteristics of patients in BRAVE-AA1 ................................... 40 Table 10. Baseline and disease characteristics of patients in BRAVE-AA2 ................................. 42 Table 11. Prior treatments for AA used among participants in BRAVE-AA1 (FAS population) .... 44 Table 12. Prior treatments for AA used among participants in BRAVE-AA2 (FAS population) .... 45 Table 13. Trial populations used for the analysis of outcomes in BRAVE-AA1 and BRAVE-AA2 46 Table 14. Critical appraisal of BRAVE-AA1 and BRAVE-AA2 ...................................................... 51 Table 15. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA1 and BRAVEAA2 (FAS population; primary censoring rule [NRI]) ..................................................................... 53 Table 16. Efficacy results for key secondary endpoints (after adjustment for multiplicity) for BRAVE-AA1 (FAS population) ....................................................................................................... 56 Table 17. Efficacy results for key secondary endpoints (after adjustment for multiplicity) for BRAVE-AA2 (FAS population) ....................................................................................................... 57 Table 18. Proportion of Patients achieving SALT≤20 at Weeks 16 and 24 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ....................................................... 59 Table 19. Proportion of Patients Achieving a SALT50 at Week 12 in BRAVE-AA1 and BRAVEAA2 (FAS population; primary censoring rule [NRI]) ..................................................................... 63 Table 20. Proportion of patients achieving an absolute SALT50 at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ....................................................... 64 Table 21. Proportion of patients achieving an absolute SALT75 at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ....................................................... 65 Table 22. Proportion of Patients Achieving an Absolute SALT≤10 at Weeks 24 and 36 in BRAVEAA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ........................................ 66 Table 23. Mean change from baseline in Skindex-16 AA domain scores (Emotions, Functioning and Symptoms) at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [mLOCF]) ................................................................................................................ 71 Table 24: Mean change in Skindex-16 AA domain scores (Emotions, Functioning and Symptoms) at Week 36 in the BRAVE-AA studies for SALT50 Responders versus Nonresponders (pooled Week 36 efficacy population; primary censoring rule [mLOCF]) .................. 73 Table 25. Mean change from Baseline in HADS-A and HADS-D Total Scores at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [mLOCF]) .................... 74 Table 26. Mean change from baseline in EQ-5D-5L health state index (UK algorithm) scores at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [mLOCF]) . 75 Table 27. Mean change from baseline in EQ-5D-5L VAS scores at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [mLOCF]) ................................................. 75 Table 28. Mean change from baseline in SF-36 PCS at Week 36 in BRAVE-AA1 and BRAVEAA2 (FAS population; primary censoring rule [mLOCF]) .............................................................. 77 Table 29. Mean change from baseline in SF-36 MCS at Week36 in BRAVE-AA1 and BRAVEAA2 (FAS population; primary censoring rule [mLOCF]) .............................................................. 78

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

==> picture [454 x 727] intentionally omitted <==

----- Start of picture text -----

----- End of picture text -----

Table 61. Drug monitoring costs for BSC .................................................................................... 126 Table 62. Costs of non-pharmacological management of the psychological burden of alopecia areata. .......................................................................................................................................... 128 Table 63. Costs of pharmacological treatment management of the psychological burden of alopecia areata ............................................................................................................................. 130 Table 64. Costs of non-pharmacological management of the psychological burden of alopecia areata in BSC ............................................................................................................................... 131 Table 65. Costs of pharmacological treatment management of the psychological burden of alopecia areata in BSC ................................................................................................................ 132 Table 66: Summary of variables applied in the economic model base case .............................. 133 Table 67: Key modelling assumptions ......................................................................................... 134 Table 68. Base case cost-effectiveness results at baricitinib PAS price (probabilistic) .............. 136 Table 69. Net health benefit ......................................................................................................... 136 Table 70. Cost-effectiveness results Scenario 1: Starting population with SALT 50–94 ............ 139 Table 71. Cost-effectiveness results Scenario 2: Starting population with SALT 95-100 ........... 140 Table 72. Cost-effectiveness results Scenario 3: Response based on SALT75 .......................... 140 Table 73. Cost-effectiveness results Scenario 4: Utilities based on EQ-5D data from baricitinib phase III studies ........................................................................................................................... 141 Table 74. Cost-effectiveness results Scenario 5: Utilities based on HADS data from baricitinib phase III studies ........................................................................................................................... 141 Table 75. Cost-effectiveness results Scenario 6: Proportion of patients on BSC drugs based on clinical expert opinion ................................................................................................................... 142

Figure 1. The JAK-STAT signalling pathway and its inhibition by baricitinib (Olumiant®) ............ 13 Figure 2. JAK-STAT signalling pathway in alopecia areata .......................................................... 17 Figure 3. Study design of BRAVE-AA1.......................................................................................... 31 Figure 4. Study design of BRAVE-AA2.......................................................................................... 33 Figure 5. Results of graphical multiple-testing procedure in BRAVE-AA1 .................................... 48 Figure 6. Results of graphical multiple-testing procedure in BRAVE-AA2 .................................... 49 Figure 7. Patient disposition to Week 36 in BRAVE-AA1 .............................................................. 50 Figure 8. Patient disposition to Week 36 in BRAVE-AA2 .............................................................. 51 Figure 9. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI]) ......................................................................................................... 54 Figure 10. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ...................................................................................... 54 Figure 11. Proportion of patients achieving SALT≤20 over time through Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI]) ............................................................................. 60 Figure 12. Proportion of patients achieving SALT≤20 over time through Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ............................................................................. 61 Figure 13. Photographs of changes in absolute SALT scores over time through Week 36 in patients treated with baricitinib in BRAVE-AA1 ............................................................................. 62 Figure 14. Proportion of patients achieving SALT≤10 through Week 36 in BRAVE-AA1 and BRAVE-AA1 (FAS population; primary censoring rule [NRI]) ....................................................... 67 Figure 15. Proportion of patients achieving SALT≤10 through Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ....................................................... 67 Figure 16. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyebrow Hair Loss™ through Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI]) ...................................................................................... 68

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 17. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyebrow Hair Loss™ through Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ...................................................................................... 69 Figure 18. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyelash Hair Loss™ through Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI]) ...................................................................................... 70 Figure 19. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyelash Hair Loss™ through Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI]) ...................................................................................... 70 Figure 20. Proportion of patients achieving SALT≤20 through Week 52 in the BRAVE-AA studies (pooled Week 52 efficacy population; primary censoring [NRI]) ................................................... 83 Figure 21. Proportion of patients achieving SALT≤10 through Week 52 in the BRAVE-AA studies (pooled Week 52 efficacy population; primary censoring [NRI]) ................................................... 84 Figure 22. Proportion of patients achieving SALT50 through Week 52 in the BRAVE-AA studies (pooled Week 52 efficacy population; primary censoring [NRI]) ................................................... 85 Figure 23. Proportion of patients with SALT≤20 response from week 52 through week 76 (randomised withdrawal population; tertiary censoring rule [NRI]) ................................................ 86 Figure 24. Maintenance of efficacy through week 76 following down-titration at week 52 (baricitinib 4 mg week 52 re-randomised responder population [BRAVE-AA2]; tertiary censoring rule [NRI]) ....................................................................................................................................... 87 Figure 25. Evidence networks based on the 4 placebo-controlled and 7 non placebo-controlled RCTs .............................................................................................................................................. 91 Figure 26. Model structure ........................................................................................................... 110 Figure 27. Convergence plot for NMB ......................................................................................... 137 Figure 28. Cost-effectiveness plane for baricitinib 4 mg compared with ‘watch and wait’ .......... 138 Figure 29. Cost-effectiveness acceptability curve for baricitinib 4mg compared with placebo ... 138 Figure 30. Tornado diagram for baricitinib 4 mg compared with ‘watch and wait’ at PAS price . 139

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Abbreviations

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.1 Decision problem, description of the technology and

clinical care pathway

B.1.1 Decision problem

The decision problem addressed within this submission is broadly consistent with the NICE final scope for this evaluation. Any differences between the decision problem addressed with this submission and the NICE final scope are outlined in Table 1.

The full anticipated marketing authorisation for baricitinib (Olumiant®) is for the treatment of adults with severe alopecia areata (AA). The indication of relevance for this submission covers the full marketing authorisation for baricitinib.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 1 The decision problem

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved Page 11 of 151

Abbreviations: AA: alopecia areata; AE: adverse event; AESI: adverse event of special interest; ClinRO: clinician reported outcome; DPCP: 2,3-diphenylcyclopropenone; EQ5D: the European Quality of Life-5 Dimensions; HADS: Hospital Anxiety Depression Scale; NRS: numeric rating scale; PRO: patient reported outcome; PUVA: psoralen plus ultraviolet A light therapy; SAE: serious adverse event; SALT: Severity of Alopecia Tool; SF-36: Short Form 36 Health Survey Questionnaire.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979] © Eli Lilly (2022). All rights reserved Page 12 of 151

B.1.2 Description of the technology being evaluated

A summary of the mechanism of action, marketing authorisation status, costs and administration requirements of baricitinib in the treatment of severe AA is presented in Table 2. The draft Summary of Product Characteristics (SmPC) is located in Appendix C.

Table 2: Technology being evaluated

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.1.3 Health condition and position of the technology in the treatment pathway

Overview and burden of the disease

• Alopecia areata (AA) is a chronic autoimmune disorder affecting the hair follicles, characterised by the sudden onset of non-scarring hair loss.[10, 11]

• • AA evolution is unpredictable, and the prognosis varies considerably depending on disease severity and duration. Patients with extensive hair loss rarely experience spontaneous regrowth of hair and tend to have a poor prognosis even on treatment.[2, 12]

• • AA affects males and females equally, and can occur at any age, though there is generally a higher incidence in younger adults.[10 ]

• • The burden of AA on patients impacts many aspects of their lives and AA is associated with a significant impact on HRQoL. AA is further associated with a variety of comorbid diseases, including psychiatric disorders and other immune diseases.[2, 13 ]

• Clinical pathway of care and unmet need • There are currently few evidence-based treatments for AA. Current clinical management for AA therefore often involves using off-label treatments, of which few have been well-evaluated in clinical trials or have consensus on their efficacy. Many treatments are also associated with unpleasant or uncomfortable side effects.[1 ]

• • Leaving AA untreated and taking a ‘watch and wait’ approach is considered by some experts as a legitimate option for many patients with AA. A substantial proportion of patients with limited patchy hair loss of short duration experience spontaneous remission, meaning that the use of off-label treatments, with their associated side effects, may not be justified.[1, 14, 15] A ‘watch and wait’ approach may also be taken initially for patients with severe disease; however, unlike those with a milder form of the disease, this subgroup of patients often has a poor prognosis and rarely experiences spontaneous remission.[1, 16 ]

• • Aside from a ‘watch and wait’ approach, various off-label treatment options may be trialled in patients with AA in an attempt to restore hair regrowth. These treatments include topical, intralesional or oral corticosteroids, topical immunotherapy such as DPCP, immunosuppressive drugs such as methotrexate, psoralen plus ultraviolet A light therapy (PUVA), minoxidil and calcineurin inhibitors.[1] These treatments are not supported by robust evidence from clinical trials, and are associated with suboptimal efficacy and many have a safety profile that limits long-term use.

• • There is a substantial unmet need for evidence-based, effective, and well-tolerated medications for the treatment of severe AA.

• • A positive recommendation for baricitinib in patients with severe AA would allow these patients to benefit from improved outcomes compared with current management, and would provide a novel, evidence-based, therapeutic option for AA.

B.1.3.1 Overview of the disease

Alopecia areata is a chronic autoimmune disease that can lead to significant, and in some cases, total, hair loss on the scalp, face and or body

Alopecia areata (AA) is a chronic autoimmune disorder affecting the hair follicles, characterised by the sudden onset of non-scarring hair loss.[10, 11] Although AA can affect any hair-bearing skin including the beard, eyebrows, eyelashes, body and limbs, the scalp is most commonly affected, with hair loss on the scalp being observed in approximately 90% of cases.[17] The clinical presentation of AA is heterogenous, with the extent of hair loss ranging from well-defined patches on the scalp to extensive or total hair loss on the scalp, face, and/or body.[2]

Disease classification is usually by extent or pattern of hair loss.[17] Patchy AA is the most common presentation of AA, which is characterised by round or oval patches of hair loss, and Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

can encompass both patients with mild and severe AA. Other types of AA include alopecia totalis (AT), referring to total or near-total hair loss on the scalp, and alopecia universalis (AU), referring to total or near-total loss of body hair. Different patterns of hair loss have also been described, such as diffuse AA where sudden thinning of hair all over the scalp is observed, rather than in patches (Table 3).[12]

Table 3. Types of alopecia areata

Abbreviations: AA: alopecia areata. Source: Lintzeri, et al. 2022.[12]

Alopecia areata is an autoimmune disease directed against the hair follicle

While the exact cause of AA is unknown, it is understood to be a multifactorial disease driven by genetic, epigenetic and environmental factors that contribute to an immune-mediated attack on hair follicles.[18] Research suggests that the pathogenesis of AA involves an interaction between lymphocytes and hair follicular cells mediated by the JAK-STAT pathway. CD8+NKG2D+ T-cells produce IFN-γ which then binds to receptors on the surface of hair follicle cells and signals via JAK1 and JAK2 to stimulate the production of interleukin [IL]-15 in the hair follicle cell (Figure 2). IL-15 is released from the hair follicle and binds to its receptor on the surface of T-cells, further stimulating the production of IFN-γ via JAK1 and JAK3 signalling and creating a positive feedback loop between the follicular cell and the CD8+NKGD2+ T-cells.[12, 19, 20] As a result, hair follicles convert prematurely from the growth (anagen) phase into the loss (telogen) phase, resulting in the hair loss that is characteristic of AA.[21]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 2. JAK-STAT signalling pathway in alopecia areata

==> picture [452 x 320] intentionally omitted <==

Abbreviations: MHC: major histocompatibility complex; TGFβ1: transforming growth factor beta 1; TCR: T cell receptor; CD8: cluster of differentiation 8; NKG2DL: natural killer group 2D ligand; NKG2D: natural killer group 2D; IL-15: interleukin-15; JAK: Janus kinase; IL-15RA: interleukin-15 alpha subunit; STAT: signal transducer of activation; P: phosphate; IFNγ: interferon gamma; IFNγR: interferon gamma receptor. Adapted from : Divito SJ, et al. 2014

Diagnosis of AA is typically determined by clinical history and physical examination and can be differentiated from other hair loss disorders

AA is typically diagnosed based on presenting features and once other causes of hair loss have been excluded.[1, 22] Aside from the patches of hair loss that are observed in AA, typical clinical features of AA that support a diagnosis include the presence of so-called “exclamation mark” hairs, that are short, broken and taper proximally.[21, 22] In addition, the pull test may be used to support an AA diagnosis, which involves the examiner grasping approximately 40–60 hairs between their thumb, index, and middle fingers and gently pulling them away from the scalp. A positive result is achieved if >10% of the hairs are pulled out, indicating hair shedding, though a negative pull test does not always rule out an AA diagnosis.[23]

While further testing is often not required beyond careful evaluation of the patients’ clinical history and thorough physical examination, additional investigations such as a trichoscopy or histopathology may sometimes be used to confirm the diagnosis. This is particularly useful for certain forms of AA, including diffuse AA.[2, 24]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

The Severity of Alopecia Tool provides an objective and standardised method to estimate the extent of scalp hair loss and describe disease severity

Determination of the severity of AA is important for optimal disease management since it informs therapeutic decision-making and can aid evaluation of clinical response and prognosis.[12] Experts are generally in agreement that the definition of AA severity should be driven by the extent of scalp hair loss, as patients with AA often report scalp hair loss as being the most bothersome symptom of AA.[25] As such, it has been suggested that the Severity of Alopecia Tool (SALT) should be used to describe disease severity, as it can provide an objective estimation of scalp hair loss in patients with AA.[1, 21] The SALT score is calculated by measuring the hair loss in each of the four areas of the scalp (left side, right side, top and back) following a physical examination or using a photograph. Within each area, the percentage of hair loss is determined independently using a visual aid, before adding the total of each area to capture the overall percentage of hair loss.[26] An absolute score of 0% indicates no hair loss, while a score of 100% indicates complete hair loss on the scalp. Given that the SALT score measures the extent of hair loss irrespective of the underlying cause, the SALT score does not assess hair loss specific to AA. In addition, the SALT score does not consider other anatomical sites beyond the scalp (e.g., beard hair), and also does not consider elements beyond hair loss, such as the impact of AA on health-related quality of life (HRQoL).[12] While further scores have been proposed to address these limitations, there is currently no consensus on their use in clinical trials or practice.[12, 27] The SALT therefore represents the only validated and standardised tool for measuring the extent of hair loss in patients with AA for which there is consensus on its use in studies and clinically.[28]

In clinical practice in the UK, SALT scores may be used by dermatologists to determine the extent of hair loss and thus, the most appropriate management option. A SALT score of ≥50 (indicating 50% or more scalp hair loss) has been used consistently in the literature and clinical practice guidelines to define severe disease, encompassing patients with extensive patchy AA, alopecia totalis and alopecia universalis.[29] Some dermatologists may combine this with an assessment of psychological burden using a HRQoL measure to give a comprehensive picture of disease severity of an individual patient.[21] However, this is not currently incorporated into clinical practice guidelines as a way in which to determine disease severity.[1, 21]

In clinical trials, SALT is also used for the assessment of scalp hair loss, and a SALT score ≥50 has also been used in this setting to define severe AA. SALT scores may also be used in clinical trials to define the efficacy of a treatment, with improvements from baseline SALT scores, as well as the proportion of patients reaching pre-defined absolute SALT scores, being used as outcomes to measure treatment success.[21, 28] For instance, in the two pivotal trials that make up the current baricitinib clinical programme for AA, BRAVE-AA1 (NCT03570749) and BRAVE-AA2 (NCT03899259), the primary efficacy endpoint is the proportion of patients achieving SALT ≤20 at week 36 (≤20% scalp hair loss, or ≥80% scalp coverage with hair).[30, 31]

The disease course of AA is variable, and patients with severe disease are more likely to have a poor prognosis

The disease course of AA is variable and often unpredictable.[1] Consistent with the autoimmune foundations of the disease, patients with AA typically experience a relapsing and remitting disease course, with as many as 85% of individuals experiencing multiple episodes of hair loss.[17] In some individuals, hair loss occurs at intervals separated by episodes of regrowth and prolonged periods of remission, while others experience areas of hair loss resolving at the same time as new patches are appearing. For other patients, there is a more persistent disease course

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

without any regrowth or further deterioration. It is estimated that between 14–25% of patients with patchy AA progress to more severe forms of AA, including alopecia totalis or alopecia universalis.[2]

Since hair follicles are not destroyed by the disease process, some patients experience spontaneous hair regrowth.[2] Early in the disease course, spontaneous hair regrowth is relatively common, with as many as 80% of patients with mild patchy AA experiencing spontaneous regrowth within 1 year.[14] However, relapses are frequent and the likelihood of recovery decreases with increasing AA severity.[2, 32] Clinical trial data indicate that when hair loss becomes extensive, it tends to be chronic and spontaneous regrowth is rare, with very few patients with severe AA experiencing full recovery (<10%).[1, 16] Moreover, patients with severe AA are more likely to have further hair loss over time.[15, 24] The main prognostic factor in patients with AA is therefore considered to be the extent of hair loss, especially at presentation.[12, 33] As such, patients with more severe AA, such as alopecia totalis or alopecia universalis, generally have a poor prognosis.[1, 16, 33] In a review of studies that evaluated recovery rates in patients with alopecia totalis or alopecia universalis with a mean follow-up of ≥5 years, it was reported that only 8.5% (32/375) of patients achieved complete recovery as an endpoint, either with or without treatment.[34] Similarly, an Italian study including 191 patients with AA reported that of those study participants with hair loss of <25% initially, 68% were disease free at follow-up compared with 32% and 8% of those who had 25–50% or >50% hair loss at presentation, respectively.[1, 15]

Aside from the extent of hair loss, other poor prognostic factors include those who have the ophiasis variant or have coexisting changes in finger- and toenails. Compared with adults, early onset of AA in children often results in both a greater degree of and progression of AA. A history of atopy and concomitant autoimmune diseases also confers poor outcomes.[1]

AA is more common in younger adults and affects men and women equally, with severe AA representing a smaller subset of all AA patients

Estimates of the general population prevalence of AA range from 0.1–0.58%.[35] Similarly, a recent UK-based population-based cohort including 4.16 million subjects in primary care reported an overall incidence rate of 0.26 per 1000 person-years and a point prevalence of 0.58% in adults in 2018.[36] Of these, clinical experts have estimated that 15–30% have a severe form of the disease, meaning that severe AA affects a smaller, but not insignificant, subset of the total number of AA patients.[37]

AA affects both males and females and can occur at any age; however, AA has been shown to be more prevalent in younger age groups. A recent UK study, which represents the largest population-based study of AA to date, found that the onset of AA peaked at age 25–29 years for both sexes.[36] Similarly, 88% of patients affected by AA are affected before the age of 40.[10] This is considerably younger than would be expected for normal age-related hair loss and is also during the age range that is arguably the most productive with respect to career, social life, and relationships, potentially exacerbating the impact of AA on this population.[10]

B.1.3.2 Burden of alopecia areata

The burden of AA on patients impacts many aspects of their lives

Patients with AA do not typically have physical symptoms that accompany their hair loss, although occasionally itching, tingling, burning or pain may occur prior to hair loss or with disease

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

activity.[22] However, hair has a range of important physiological functions, including acting as a physical barrier to UV rays and protecting the eyes from debris. Therefore, some patients with AA experience a range of physical challenges that may impact quality of life, including excessive sneezing and/or a runny nose due to nasal hair loss, as well as sunburn from the lack of bodily hair.[38] Loss of the protective function of eyebrows and eyelashes also presents physical challenges that are not easily compensated for, due to water, sweat, and other debris getting into the eyes.[38]

Beyond its physical functions, hair also has psychosocial importance in society, as it plays a role in both sexual and social communication. It is therefore well recognised that AA can have a profound psychological and psychosocial impact on individuals.[2, 39, 40] For instance, healthy hair is generally considered an indication of youth and vigour, and plays a critical role in identity and self-worth.[25, 41] In addition, certain cultures consider hair sacred, with some religions viewing uncut hair as a sign of devotion.[41] These strong cultural meanings behind hair have been highlighted by AA patients as a factor that contributes to the significant psychological impact of AA. For example, women expressed seeing hair as a mark of their femininity, and therefore likened losing their hair to having a mastectomy.[42] The hair loss in AA can therefore negatively impact self-esteem, body image and confidence, leading to feelings of trauma, shock and loss of identify.[11, 42] People with AA also often report feeling judged and may experience stigma due to their hair loss; one study reported that the general population associated those with severe hair loss as being sick, unattractive and even dirty.[43] Patients with AA therefore often report the highly visible manifestations of AA as the most troublesome aspect of AA and the primary cause of their distress.[38, 40]

Given the feeling of social judgment and stigma experienced by people with AA, attempts at concealment are common, including the use of wigs, fake eyelashes and eyebrows. However, using these methods of concealment can produce feelings of inauthenticity, shame, and anxiety.[42] Furthermore, some patients report negative experiences associated with wig-wearing, including itchiness, discomfort, the fear of it being dislodged, and worry about the reactions of others with respect to them wearing a wig.[11]

The unpredictable disease course is another aspect of AA that can be very stressful for patients, and can contribute to the psychological burden of the disease.[44] In a qualitative study, patients reported that the unpredictable and often rapidly alternating cycles of hair loss and regrowth are particularly disturbing aspects of the disease.[42] Some patients may find it particularly difficult to cope with relapse, or if new bald areas occur as others improve.[1, 45] In addition, patients have indicated that the pattern of hair loss, with randomly distributed visible patches of hair loss, is particularly challenging. The poor prognosis associated with the disease may also be particularly difficult for patients to cope with and can lead to feelings of hopelessness.

AA is associated with a significantly reduced HRQoL

Given the psychological and emotional distress caused by the sometimes extensive hair loss, many patients with AA experience a reduced health-related quality of life (HRQoL).[46-49] Various different instruments have been applied to AA to measure its impact on HRQoL, including generic measures such as SF-36.[2] For instance, a meta-analysis of studies employing the SF-36 instrument reported that patients with AA experience a significant impairment in HRQoL compared to age- and sex-matched controls. HRQoL impairment was most pronounced across the role-emotional, mental health and vitality domains (P<0.001), indicating poor social functioning, higher psychological distress, and diminished energy levels.[47]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Given the mono-symptomatic nature of the hair loss in AA and the fact many patients with AA are otherwise healthy, other more dermatology-specific measures are also widely used in order to better define the factors affected HRQoL, such as the Skindex-16 and the Dermatology Life Quality Index (DLQI).[2, 16] Using these instruments and others, studies consistently demonstrate the poor HRQoL experienced by patients with AA, which can be similar to those reported in patients with other chronic skin disease, including atopic dermatosis and psoriasis.[46-49] For instance, a systematic literature review (SLR) investigating the effects of AA on HRQoL showed that people with AA had consistently low HRQoL scores. These scores, measured using a range of different instruments (DLQI, Skindex, SF-36), were comparable to other chronic skin diseases, despite the lack of physical symptoms beyond hair loss that are known to reduce HRQoL, such as itching and sleep disturbance.[48] Similarly, an observational cross-sectional study conducted in a large cohort of patients with a range of dermatological conditions (n=4010) across 13 European countries demonstrated that EQ-5D VAS scores were comparable in people with AA (n=33, mean: 69.7%; SD:18.1) and those with AD (n=177, mean: 66.0%; SD: 19.0]) or psoriasis (n=682, mean: 65.6%; SD: 20.0) and lower than healthy controls (n=1359, mean: 82.2%; SD: 15.5). The authors also noted that the EQ-5D VAS scores of <70 associated with several dermatological conditions, including AA, were comparable to published health state estimates for chronic diseases such as diabetes mellitus, cardiovascular disease, anxiety, cancers, and liver disease.[50]

Several studies also demonstrate the greater quality of life impairment in patients with more severe disease.[47-49, 51] An SLR of 21 studies conducted by Rencz et al. (2016) identified that greater scalp involvement, disease recurrence and longer treatment duration all negatively impacted HRQoL.[47] Similarly, in a recent Europe-wide study, patients with AA had worse HRQoL (measured by DLQI) compared to age-, sex- and comorbidity-matched controls and patients with androgenic alopecia (AGA) (p=0.022), with greater DLQI score impairment in those with the greatest AA severity.[39]

AA is associated with psychological burden and a variety of comorbid diseases

In addition to the quality of life impairment reported in patients with AA, psychological stress levels, frequency of psychiatric disease and levels of psychiatric symptoms are typically higher in adult patients with AA than controls.[2, 11, 13] As such, it is estimated that there is a 66%–74% lifetime prevalence of psychiatric disorders in AA patients, with a 39% lifetime prevalence of depression and a 39%–62% prevalence of generalised anxiety disease.[10] In the UK, a study including 338 patients with alopecia (mixed-severity AA, n=279) reported clinically significant levels of social anxiety (37.5%), anxiety (35.5%) and depression (29%).[11] Similarly, in a population-based study based in UK primary care, adults with newly diagnosed mixed-severity AA (n=5,435) had significantly higher background prevalence of depressive episodes (DE), recurrent depressive disorder (RDD) and anxiety disorder (AD) (DE 19.4%, RDD 12.3%, AD 16.6%) compared to matched controls (n=21,740; DE 14.7%, RDD 8.6%, AD 12.9%). This study also found that those with AA were more likely to go on to develop new-onset DE and AD (adjusted hazard ratio [95% CI] 1.38 [1.13, 1.69]), RDD (1.30 [1.04, 1.62]), and AD (1.33 [1.09, 1.63]). Higher rates of antidepressant prescribing were also seen in people with AA.[13] These conditions may be primary disorders that manifest themselves as medical problems associated with AA, or could result from the chronic, relapsing nature of AA and its negative effect on a person’s appearance.[39] Similarly, during an observational study among patients with a range of dermatological conditions (n=4010), AA was associated with significantly worse anxiety and depression compared with controls (OR [95% CI] 4.19 [2.0, 8.9]; p<0.05), prompting the authors to suggest there may be a need for psychiatric support in such individuals.[50] Another observational cross-sectional study in 17 European countries found that patients with Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

dermatological conditions, including alopecia, had a more than eleven-fold increased chance of having symptoms of body dysmorphic disorder compared with controls, a common psychiatric disorder associated with high costs for healthcare systems.[51] In severe AA patients with a longer disease duration, rates of depression and anxiety may be even greater than estimates in mixedseverity populations, given that greater disease severity is associated with greater impairment in quality of life.[47-49]

There have also been reports of suicidal ideation in people with AA, translating into an elevated mortality risk associated with intentional self-harm/psychiatric disease among patients with AA compared with control subjects, with the risk being particularly elevated among individuals with alopecia totalis (AT) or alopecia universalis (AU).[52, 53]

Besides psychiatric disorders, it has been reported that AA is associated with a variety of other comorbidities, including autoimmune or atopic diseases, which can also contribute to the overall burden of the disease (Table 4).[12, 17, 27] Other comorbidities that appear to be more prevalent among AA patients include vitamin D deficiency, iron-deficiency anaemia, metabolic syndrome, and infection with Helicobacter pylori.[12, 27]

Table 4. Comorbid diseases commonly reported in patients with AA.

Abbreviations: AA: alopecia areata. Source: Lintzeri, et al. 2022; Lee, et al. 2019a.[12, 27]

AA negatively impacts employment and relationships

The impact of AA on working life has been explored in several studies, which have shown that people with AA are treated differently by others because of their disease, or that they have limited their professional lives due to their condition.[40, 43] In a UK population-based study by Macbeth et al. (2022) that included 5435 people with AA and 21,740 matched controls, certificates for time off work were issued more frequently significantly (p<0.001) to people with AA (13.0% within a year of diagnosis) compared to matched controls (7.9%). Similarly, people with AA were more likely to have a record of unemployment in the year after diagnosis (1.3% of AA cases vs 0.6% of matched controls).[13]

The negative effect of AA on relationships with friends, family, or a romantic partner has also been well documented. Patients with AA may feel that they are unable to have a romantic relationship due to their hair loss, or may experience the end of a romantic relationship because their partner was unable to cope with the hair loss.[42] Further to this, people with AA may also withdraw from social situations frequently; a survey of individuals with mixed-severity AA found that avoiding social activities (62%) and reducing interactions with friends (54%) were both Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

common after a first episode of AA, highlighting the impact of AA on all aspects of an individual’s social life.[54]

B.1.3.3 Clinical pathway of care

Current clinical management of AA usually relies on the use of off-label treatments that are not supported by robust evidence

Following on from recent approvals from the European Medicines Agency (EMA) and Food and Drug Administration (FDA),[55, 56] baricitinib will be the first MHRA-approved therapy specific to adult patients with severe AA, and the only licensed therapy for any form of AA other than intralesional corticosteroids. The current clinical management for AA in the UK therefore mostly relies on using off-label treatments; however, these options are not supported by robust evidence and their reported efficacies are low. Clinical practice guidelines specific to the UK include the NICE Clinical Knowledge Summary for AA, and the British Association of Dermatologists’ (BAD) guidelines for the management of alopecia areata, published in 2012.[1, 29] However, these are limited in their treatment recommendations and highlight a lack of high-quality evidence for treatments in AA, with the majority of evidence based on studies with small sample size and short follow-up times.

Initial management of AA in the UK is typically based on the severity of hair loss on the scalp and the priorities of AA treatment generally include to:[29]

• Arrest the progression of hair loss and induce hair growth

• Improve patients’ HRQoL

• Limit the adverse events related to therapy

Treatment of mild AA

In the UK, patients with mild AA (<50% scalp hair loss) commonly receive no treatment immediately after diagnosis and instead undergo a period of ‘watch and wait’, where management involves the provision of reassurance and advice on the nature and course of AA alone.[1, 29] This is based on the fact that up to 80% of patients with limited patchy hair loss experience spontaneous regrowth within a relatively short period of time, meaning that the use of off-label treatments, with their associated side effects, may not be justified.[1, 14, 15] While waiting for their hair to regrow, these patients can benefit from advice on cosmetic options to conceal hair loss, such as wigs or protheses, as this can help the patient cope with the psychological impact of the disease.[1, 29] Patients may be eligible for a free or reduced cost wig on the NHS.[29, 57] Psychological support may also be valuable for some patients, which can include contact with patient support organisations, such as the National Alopecia Areata Foundation and Alopecia UK, as well as professional support from a clinical psychologist or other practitioner skilled in helping patients cope with their mental health.[1, 2, 44]

Although many patients with mild AA are likely to spontaneously remit without intervention, it may be preferrable for some patients to initiate treatment immediately after diagnosis. In these patients, treatment typically involves the use of either potent topical steroids or intralesional corticosteroids. While topical corticosteroids may advance regrowth in some patients with mild AA, evidence supporting the effectiveness of topical steroids is generally limited and often conflicting.[1, 45] Similarly, while intralesional corticosteroids (the only other licensed treatment for AA) are generally considered the most effective treatment for mild patchy AA, evidence is limited, especially in the long term.[1, 2] This treatment involves injecting corticosteroids directly into areas Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

of hair loss (~1 injection per square centimetre [cm²]), resulting in high drug concentrations at the lesion site.[58] Both of these treatments may also be trialled in patients with severe AA, though the benefits appear to be particularly low in this subgroup, as discussed below.[1, 16]

There is a substantial unmet need for evidence-based, effective, and well-tolerated medications for the treatment of severe AA

Given the lack of evidence-based treatment options for AA and the often uncomfortable and unpleasant side effects of treatment, leaving AA untreated and taking a ‘watch and wait’ approach may also be used initially for patients with severe AA. However, unlike those with a milder form of the disease, this subgroup of patients often has a poor prognosis and rarely experiences spontaneous remission.[1, 16]

Aside from a ‘watch and wait’ approach, various other mostly off-label treatment options may be trialled in patients with severe AA including topical, intralesional or oral corticosteroids, topical immunotherapy such as DPCP, immunosuppressive drugs such as methotrexate, psoralen plus ultraviolet A light therapy (PUVA), or minoxidil and calcineurin inhibitors.[1] Patients may also benefit from supportive care options such as psychological support and cosmetic concealment of hair loss through the use of wigs may be used to help these patients cope with the potentially significant physiological and psychosocial impact of the condition, which may be particularly pronounced among patients with severe AA.[1, 47-49, 51] Unfortunately, the quality of wigs can vary and some patients may also feel self-conscious about wearing a wig for fear of being discovered.[1, 57]

While topical corticosteroids and minoxidil may advance regrowth in some patients with mild AA, evidence supporting the effectiveness of the treatments is generally limited and often conflicting, especially in patients with severe disease.[1, 45] Therefore, these patients typically fail to achieve a sustained and clinically meaningful response, if any, with these treatments.[1, 2] Evidence for intralesional corticosteroids is similarly limited.[1] Prospective studies have shown that the intralesional injections of corticosteroids, in the form of triamcinolone acetonide (5–10 mg/ml) can stimulate tufts of hair regrowth at 60–67% of injection sites.[44] However, this effect is temporary, meaning that further injections are required every 4 to 6 weeks in order to achieve a more sustained response.[1] Frequent injections are resource intensive and can also lead to common complications such as skin atrophy and hypopigmentation, and may also be uncomfortable and unpleasant for patients during administration.[1, 2] In addition, due to the number of injections and drug volumes that would be required for large surfaces, intralesional steroids are not considered feasible beyond 20% hair loss.[33, 59]

PUVA, calcineurin inhibitors and continuous and pulsed systemic oral corticosteroids have also been used to treat AA, though due to their potentially serious side effects and inadequate evidence of efficacy, the risks associated with these treatments may not be justified in many patients.[1] For instance, in the only placebo-controlled study of systemic corticosteroid use in AA, 8 of 23 eight patients with >40% hair loss receiving a weekly single dose of prednisolone had substantial (>31%) hair regrowth compared with none in the placebo group, though this was not statistically significant. However, within three months, 25% of the responders had relapsed.[60] Therefore, while a response may be observed in some patients, continued treatment is usually needed to maintain hair growth and avoid relapse. Furthermore, the observed response is in many cases insufficient to justify the known side effects of systemic corticosteroid use, especially those observed with long-term treatment, such as glaucoma, hypertension, osteoporosis and Cushing’s syndrome.[1, 44, 61] Similarly, the use of PUVA is characterised by high rates of relapse

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

and unacceptably high cumulative UVA doses, and its use is therefore not recommended in the BAD clinical practice guidelines.[1]

Contact immunotherapy is the best-documented treatment in severe AA, though there are no RCTs comparing this intervention with placebo. The aim of treatment is to induce low grade allergic contact dermatitis by initially sensitising the patient, and then applying very weak concentrations of a contact allergen directly to the scalp once a week.[1, 44] Commonly used contact allergens include 2,3-diphenylcyclopropenone (DPCP) and squaric acid dibutyl ester (SADBE), though the former is usually the agent of choice.[1] A review of all the published studies of contact immunotherapy concluded that although the response rates vary widely (9–87%), in general, 20–30% patients achieve a worthwhile response, such as sufficient regrowth to enable patients to manage without a wig.[62] However, relapse rates on maintenance regimens or following discontinuation are high.[1, 12, 63] Topical immunotherapy is generally not well tolerated, and can induce severe contact dermatitis in patients and, sometimes, in the provider. It also involves multiple visits to hospital over several months.[1] Contact immunotherapy is also limited to selected specialist centres in the UK, which limits its role in AA management.[1]

Unmet need and positioning of baricitinib in the clinical pathway of care

The expected eligible patient population for baricitinib in UK clinical practice is adult patients with severe AA. This population is in line with the anticipated license indication for baricitinib in the UK and the eligibility criteria for the BRAVE-AA1 and BRAVE-AA2 trials.[30, 31]

Despite the significant burden associated with AA, there are currently few evidence-based treatment options for patients with AA. The current management of AA therefore chiefly relies on ‘watch and wait’ or off-label treatments that are not supported by robust evidence from clinical trials.[1] While some patients respond to existing therapies initially, most experience relapse after stopping treatment and most current modalities cannot be used in the long term due to adverse events. In addition, current treatments can be uncomfortable for the patient, are time-consuming and have limited availability. The efficacy of these treatments is also reported to be particularly low in patients with severe AA.[1, 21, 33] Existing treatments for AA are therefore insufficient for patients with severe disease due to the lack of robust efficacy evidence that often cannot outweigh the poor long-term safety and tolerability.[1] As such, many patients with severe AA must rely on management options such as concealment of hair loss and psychological support to cope with the potentially significant physiological and psychosocial impact of the condition, even if they still want to be treated.[1]

The results of the pivotal Phase III clinical trials for baricitinib, BRAVE-AA1 and BRAVE-AA2, demonstrate that baricitinib is an efficacious new treatment for AA with an acceptable safety profile, including with longer-term use up to 52 weeks. Among patients with severe AA, baricitinib significantly improved hair regrowth on the scalp, eyebrows and eyelashes, with some responders experiencing complete hair regrowth within 36 weeks. In addition, Week 52 data indicate that hair regrowth continues to increase among responders beyond 36 weeks, demonstrating that some patients may experience further clinical benefit from baricitinib beyond this timepoint. Furthermore, Week 76 data suggest that efficacy is maintained over time in most of the patients who have reached SALT≤20 by Week 52. Patients who received baricitinib also experienced greater reductions in Skindex-16 AA and HADS scores compared with placebo, indicating an improvement of the HRQoL and psychological burden associated with AA.[30, 31]

A positive recommendation for baricitinib in this setting would therefore allow patients to benefit from improved outcomes compared with current management, reducing the significant and

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

negative burden of the disease on patients’ lives. This would also provide the first evidencebased therapeutic option to address the significant unmet need for an effective and tolerable treatment option for patients with severe AA, by representing the first evidence-based treatment specific to AA.

B.1.4 Equality considerations

No equality issues related to the use of baricitinib in this indication have been identified or are foreseen.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2 Clinical effectiveness

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.1 Identification and selection of relevant studies

A systematic literature review (SLR) was conducted to identify all relevant clinical evidence on the efficacy and safety of baricitinib and comparators for the treatment of AA. The original SLR was conducted in July 2021 and was subsequently updated in February 2022. In total, the updated SLR identified 47 unique studies, with evidence generated from 13 RCTs and 34 observational studies. One randomised clinical trial investigating the safety and efficacy of baricitinib, the phase II portion of BRAVE-AA1, was identified during the SLR; the Phase III portion of the BRAVE-AA1 study will be presented in detail throughout Section B2. The remaining identified studies investigated various baricitinib comparators, the details of which will be discussed in greater detail in Section B.2.9. Full details of the SLR search strategy, study selection process and results can be found in Appendix D. A further study, investigating the safety and efficacy of baricitinib, BRAVE-AA2, was not captured in the SLR, however this is presented in detail throughout Section B2.

B.2.2 List of relevant clinical effectiveness evidence

One publication was identified in the SLR that provided clinical evidence for the efficacy and safety of baricitinib for the treatment of severe AA. King et al. (2021) reports the Phase II portion of the randomised, double-blind, placebo-controlled trial BRAVE-AA1 [NCT03570749].[64] Evidence for the efficacy and safety of baricitinib in severe AA is further provided by King et al. (2022) reporting the Phase III portion of the BRAVE-AA1, and the identically-designed Phase III study BRAVE-AA2 [NCT03899259].[30, 31] Given the availability of these Phase III data, the Phase II portion of the BRAVE-AA1 trial has not been considered further within this submission. Overviews of these Phase III RCTs are provided in Table 5.

The patient populations for BRAVE-AA1 and BRAVE-AA2 were adult patients with severe, with limited permitted concomitant medications allowed. This population is in line with the population of relevance for this evaluation; adult patients with severe AA. As such, a pooled population of the patients from BRAVE-AA1 and BRAVE-AA2 informs the base case economic analysis.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 5. Clinical effectiveness evidence

Abbreviations: AA: alopecia areata; ClinRO: clinician reported; HRQoL: health related quality of life; EQ-5L-5L: 5-level EuroQol 5 Dimensions; HADS: hospital anxiety and depression score; PRO: patient reported outcome; SALT: severity of alopecia tool; SF-36: medical outcomes study 36-item short form health survey. Source : BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.3 Summary of methodology of the relevant clinical effectiveness evidence

B.2.3.1 Summary of trial methodology

BRAVE-AA1 and BRAVE-AA2 are ongoing, multicentre, double-blind, randomised, placebocontrolled studies to determine the safety and efficacy of baricitinib in adult patients with severe AA. BRAVE-AA1 was an adaptive phase II/III study. The Phase II portion of the study was designed to determine the doses of baricitinib for use in the phase III trials; doses of 1 mg, 2 mg and 4 mg were evaluated. Based on the findings from the phase II portion of BRAVE-AA1, the 2 mg and 4 mg doses were selected for further investigation in the Phase III portion of BRAVE-AA1 and the Phase III BRAVE-AA2 study. The current submission will focus on the Phase III results only as the data from patients enrolled in the phase II portion of BRAVE-AA1 are not included in the efficacy analyses of the phase III part of the trial.

The patient populations for BRAVE-AA1 and BRAVE-AA2 were adult patients with severe AA. This was defined as patients with a current AA episode of more than 6 months duration prior to screening (Visit 1), with at least 50% scalp hair loss at screening and baseline (Visits 1 and 2) and no spontaneous improvement (no more than a 10-point spontaneous reduction in Severity of Alopecia Tool [SALT] score) over the past 6 months. The use of concomitant medications for the management of AA was generally prohibited throughout the trial, with some exceptions detailed within the study protocols.

The study design of BRAVE-AA1 and BRAVE-AA2 is shown in Figure 3 and Figure 4, respectively. Before Week 52, the study designs of BRAVE-AA1 and BRAVE-AA2 were largely identical. Following a screening period between 3 and 35 days prior to Visit 2 (Week 0) patients were randomised in a double-blinded fashion to once-daily treatment with placebo, 2 mg baricitinib or 4 mg baricitinib (2:2:3) Once patients completed Week 36, they entered the longterm extension period and, through to Week 52, either remained on baricitinib regardless of response, remained on placebo if they had achieved SALT≤20 at Week 36, or in the case of nonresponse in placebo-treated patients were rescued to baricitinib. From Week 52 onward, the study designs for BRAVE-AA1 and BRAVE-AA2 differed. In BRAVE-AA1, baricitinib-treated (2 mg or 4 mg) patients who achieved SALT≤20 at Week 52 were eligible to participate in a randomised withdrawal sub-study if they had remained on the same dose of baricitinib from initial randomisation. In BRAVE-AA2, patients treated with 4 mg baricitinib who achieved SALT≤20 at Week 52 were eligible to participate in a randomised down-titration sub-study if they had remained on baricitinib 4 mg from initial randomisation. Further information on the predefined criteria for continuation into the long-term extension and bridging extension is provided in the study protocols.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 3. Study design of BRAVE-AA1

==> picture [618 x 406] intentionally omitted <==

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979] © Eli Lilly (2022). All rights reserved Page 31 of 151

Footnotes:[a ] Placebo responders stayed on placebo for remainder of the trial, even if relapse was observed later.[b] Patients with SALT ≤20 who stayed on the same dose of baricitinib from week 0 were randomised to stay on current baricitinib dose, or transitioned to placebo.[c] Responders participating in randomised withdrawal who experienced >20-point absolute worsening in total SALT score after week 52 were retreated with baricitinib dose to which they were originally randomised if they were randomised to placebo at week 52, OR continued to receive same dose of baricitinib if they were randomised to remain on baricitinib at week 52.[d] Non-responders 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 achieved SALT ≤20 before week 52.[e] Never responders (never achieved SALT ≤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] Non-responders at week 52 AND week 76 were automatically discontinued at week 76 unless they had a ≥2-point improvement from baseline in ClinRO measures for EB or EL hair loss. Abbreviations: PBO: placebo; PTFU: post-trial follow up; QD: once daily; W: week. Source : BRAVE-AA1 Clinical Study Report.[65]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 4. Study design of BRAVE-AA2

==> picture [610 x 400] intentionally omitted <==

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979] © Eli Lilly (2022). All rights reserved Page 33 of 151

Footnotes:[a ] Placebo-treated patients not eligible for rescue to baricitinib at week 36 (due to spontaneous remission) were rescued to baricitinib if they were non-responders at week 52, OR if they experienced loss of treatment benefit after week 52.[b] Patients randomised to baricitinib 2 mg at week 0 were rescued to the 4-mg dose if they were nonresponders at week 52, OR were responders at week 52 but experienced a >20-point worsening in SALT score after week 52.[c] Responders in the baricitinib 4-mg group (SALT ≤20 who stayed on 4 mg from week 0) were randomised to either stay on 4 mg OR transition to 2 mg.[d] Responders participating in the randomised down-titration who experienced a loss of treatment benefit after week 52 were re-treated with baricitinib 4 mg if they were randomised to the 2-mg dose at week 52, OR continued to receive baricitinib 4 mg if they randomised to remain on the 4-mg dose at week 52.[e ] At week 52, non-responders (SALT >20) in the baricitinib 4-mg group since baseline who achieved SALT ≤20 before week 52 remained on 4 mg.[f] Never responders (never achieved SALT ≤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] Non-responders at week 52 AND week 76 were automatically discontinued at week 76 unless they had a ≥2-point improvement from baseline in ClinRO measures for EB or EL hair loss. Abbreviations: PBO: placebo; PTFU: post-trial follow up; QD: once daily; W: week. Source : BRAVE-AA2 Clinical Study Report.[66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979] © Eli Lilly (2022). All rights reserved Page 34 of 151

The primary outcome of both trials was to evaluate whether 2 mg or 4 mg baricitinib is superior to placebo, as measured by the proportion of patients in each treatment group achieving SALT≤20 after Week 36 of treatment. Key secondary endpoints measured scalp hair regrowth at Week 12, 16, 24 and 36 using various other SALT score thresholds. Hair regrowth was also measured at Week 36 using ClinRO for eyebrow and eyelash hair loss and patients reported outcomes (PRO scalp hair assessment score). Other secondary outcomes included HRQoL outcomes, including Skindex-16 AA domain scores and Hospital Anxiety and Depression Scale (HADS). Safety outcomes included AEs, SAEs and TEAEs by Week 36.

A summary of the methodology of BRAVE-AA1 and BRAVE-AA2 is presented in Table 6.

Table 6: Summary of methodology for BRAVE-AA1 and BRAVE-AA2

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes:[a] Copyright issues prevented Skindex-16 from being available at the start of the phase III portion of BRAVE-AA1, therefore, it could not be given to the entire intention-to-treat set. For this reason, it was designated as an exploratory outcome. For BRAVE-AA2, it was available from the start of the study, so it was included as a secondary non-gated outcome measure.

Abbreviations: AA: alopecia areata; BMI: body mass index; AA-IGA: alopecia areata investigator global assessment; ClinRO: clinician reported; HMG CoA: 3-hydroxy-3-methylglutaryl coenzyme A; HRQoL: health related quality of life; EQ-5L-5L: 5-level EuroQol 5 Dimensions; HADS: hospital anxiety and depression score; JAK: Janus kinase; PRO: patient reported outcome; SALT: severity of alopecia tool; SF-36: medical outcomes study 36-item short form health survey.

Source : BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.3.2 Outcome definitions

All outcome definitions were consistent across BRAVE-AA1 and BRAVE-AA2 trials and are summarised in Table 7.

Table 7. Definitions of outcomes used in BRAVE-AA1 and BRAVE-AA2

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Abbreviations: AA: alopecia areata; ClinRO: clinician reported outcomes; HRQoL: health related quality of life; EQ-5L-5L: 5-level EuroQol 5 Dimensions; HADS: hospital anxiety and depression score; PRO: patient reported outcome; SALT: severity of alopecia tool; SF-36: medical outcomes study 36-item short form health survey; VAS: visual analogue scale. Source : BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

SALT Score Interpretation

BRAVE-AA1 and BRAVE-AA2 employed both absolute and subscript SALT scores as outcomes to measure hair regrowth following treatment with baricitinib. The interpretation of these types of SALT scores is explained in Table 8. An absolute SALT score represents a percentage of scalp hair loss, which can also be expressed as a percentage of scalp coverage with hair, whereas a subscript of the SALT score represents a percentage change from baseline.

Table 8. Interpretation of the SALT score

Abbreviations: SALT: Severity of Alopecia Tool.

B.2.3.3 Baseline characteristics of study participants

Demographic and Disease Characteristics

Patient demographics were similar between the BRAVE-AA1 and BRAVE-AA2 trials and were generally well balanced across treatment groups within each study. In addition, baseline disease characteristics and geographic region distribution did not differ substantially between the two studies and were well balanced across treatment groups (Table 9, Table 10).

Patients enrolled in BRAVE-AA1 and BRAVE-AA2 had a mean age of 37.5 years, with women representing a slightly higher proportion of the patient population across the studies compared with males (61% vs 39%). The median SALT score was 96 across the study populations, and the mean disease episode duration was 3.9 years. The mean duration from the first onset of AA diagnosis was 12.2 years, and over half of all patients enrolled in the studies had very severe AA at baseline (defined as SALT 95–100), with *** of patients having investigator-reported alopecia universalis (AU).[67]

Across the two studies, 69% of patients had significant or complete eyebrow hair loss at baseline, and 58% had significant or complete eyelash hair loss at baseline, as measured by ClinRO measures for eyebrow and eyelash hair loss scores of 2 or 3. Approximately 38% of patients reported an atopic background, defined as a medical history of, or ongoing atopic dermatitis, allergic rhinitis, allergic conjunctivitis, or allergic asthma.[67]

Table 9. Baseline and disease characteristics of patients in BRAVE-AA1

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes:[a] Disease characteristics data based on observations in n=188 patients.[b ] Includes South Korea.[c ] Includes Mexico.[d ] Atopic background is defined as having an ongoing or a medical history of atopic dermatitis, allergic rhinitis, allergic conjunctivitis, or allergic asthma.[f ] Copyright issues prevented Skindex-16 from being available at the start of the phase III portion of BRAVE-AA1, so it could not be administered to the entire intention-to-treat set (PBO N=119; 2 mg baricitinib N=111; 4 mg baricitinib N=175).

Abbreviations: AA: alopecia areata; AU: alopecia universalis; BMI: body mass index; ClinRO: clinician reported outcomes; HADS: hospital anxiety and depression score; PBO: placebo; PRO: patient reported outcome; SALT: severity of alopecia tool; SD: standard deviation.

Source: BRAVE-AA1 Clinical Study Report; King et al. 2022; EPAR.[65, 67, 68]

Table 10. Baseline and disease characteristics of patients in BRAVE-AA2

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes:[a] One patient in the placebo group in BRAVE-AA2 was inadvertently enrolled with a SALT score of 32; therefore, this patient did not fall within the severe or very severe SALT categories.[b ] Includes Japan, China, Taiwan and South Korea.[c] Includes Australia, Brazil, Argentina and Israel.[d] Atopic background is defined as having an ongoing or a medical history of atopic dermatitis, allergic rhinitis, allergic conjunctivitis, or allergic asthma Abbreviations: AA: alopecia areata; AU: alopecia universalis; BMI: body mass index; ClinRO: clinician reported outcomes; HADS: hospital anxiety and depression score; PBO: placebo; PRO: patient reported outcome; SALT: severity of alopecia tool; SD: standard deviation. Source: BRAVE-AA2 Clinical Study Report; King et al. 2022; EPAR.[66-68]

Prior and concomitant therapy

Approximately *** of the patients in the BRAVE-AA1 (Table 11) and BRAVE-AA2 (Table 12) trials reported using prior medications to treat their AA. Overall, more than *** of patients had used topical immunotherapy and **** had used systemic immunosuppressant or immunomodulator therapy, the most common of which were corticosteroids. JAK inhibitors had been used by almost ** of patients, however it should be noted that prior inadequate response to JAK inhibitors was an exclusion criterion for the trials. The high proportion of patients who had received prior medications highlights the inadequacy of current treatment options in maintaining a longer-term hair regrowth and reflects the unmet need for novel treatment options in patients with severe AA.

A washout of systemic and topical treatments for AA was incorporated before randomisation to minimise the confounding effects of prior treatment. Durations of required washouts took into consideration that a delay of several weeks may be observed between treatments and regrowth of hair in patients with AA.

Only very few concomitant medications and procedures for the treatment of AA were permitted during BRAVE-AA1 and BRAVE-AA2, and only **** of the enrolled patients were reported to have used such therapies.

Table 11. Prior treatments for AA used among participants in BRAVE-AA1 (FAS population)

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes:[a] Patients may have had ˃1 prior therapy; therefore, the n in each group may not add up to the total N for the column.[b] Topical therapies excluding IMT.[c ] All immunosuppressant and immunomodulator agents including CS, JAK inhibitors, and others.[d] Procedures include cryotherapy, micro-needling, and platelet-rich plasma injections. Abbreviations: IMT: immunomodulatory therapy; JAK: Janus kinase; PBO: placebo. Source: BRAVE-AA1 Clinical Study Report; EPAR.[65, 68]

Table 12. Prior treatments for AA used among participants in BRAVE-AA2 (FAS population)

Footnotes:[a] Patients may have had ˃1 prior therapy; therefore, the n in each group may not add up to the total N for the column.[b] Topical therapies excluding IMT.[c ] All immunosuppressant and immunomodulator agents including CS, JAK inhibitors, and others.[d] Procedures include cryotherapy, micro-needling, and platelet-rich plasma injections. Abbreviations: IMT: immunomodulatory therapy; JAK: janus kinase; PBO: placebo. Source: BRAVE-AA2 Clinical Study Report; EPAR.[66, 68]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.4 Statistical analysis and definition of study groups in the relevant clinical effectiveness evidence

B.2.4.1 Analysis sets

The analysis sets used in the analysis of the BRAVE-AA1 and -AA2 are presented in Table 13.

Table 13. Trial populations used for the analysis of outcomes in BRAVE-AA1 and BRAVEAA2

Abbreviations: AGA: androgenetic alopecia; FAS: full analysis set; GCP: good clinical practice; mFAS: modified full analysis set; PPS: per-protocol set.

Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65-67]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.4.2 Statistical methods

Treatment comparisons of discrete efficacy variables were made using a logistic regression analysis. In the case when logistic regression model did not produce statistical results due to sparse data, Fisher exact test were used. Patients were generally considered non-responders for the non-responder imputation- (NRI-) based analysis if they did not meet clinical response criteria or if they discontinued study or study treatment at any time prior to the time point of interest for any reason.

Treatment comparisons of continuous efficacy and health outcome variables were be made using analysis of covariance (ANCOVA). Type III tests for LSMs were used for statistical comparison between treatment groups

Fisher exact test were used for all adverse events, discontinuation, and other categorical safety data. Continuous vital signs and laboratory values were analysed by an analysis of covariance (ANCOVA) with treatment and baseline values.

B.2.4.3 Censoring

In the phase III BRAVE-AA1 and in BRAVE-AA2 trials, efficacy data were analysed according to a prespecified censoring rule, where data collected after permanent study drug discontinuation or data collected during remote visits due to the COVID-19 pandemic were censored. This applied to all efficacy analysis populations up to Week 52. Missing data were imputed using nonresponder imputation (NRI) for categorical endpoints and modified last observation carried forward (mLOCF) for continuous endpoints.

For endpoints tested outside of the graphical scheme, the statistical significance on the categorical variables concluded throughout are all based on the p-value from the logistic regression model (p-valueb) with a p-valueb ≤0.05 deemed significant.

B.2.4.4 Multiplicity

In the phase III portion of BRAVE-AA1 and in BRAVE-AA2, multiplicity-controlled analyses were performed on the primary and key secondary endpoints using a graphical multiple-testing procedure to control the overall family-wise type I error at a two-sided 5% significance level. Adjustments for multiplicity were not performed on other endpoints. The results of the graphical multiple-testing procedures in BRAVE-AA1 and BRAVE-AA2 are shown in Figure 5 and Figure 6 below.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 5. Results of graphical multiple-testing procedure in BRAVE-AA1

==> picture [454 x 374] intentionally omitted <==

Abbreviations : ClinRO, clinician-reported outcome; EB, eyebrow; EL, eyelash; PCFB, percent change from Baseline; PRO, patient-reported outcome; SALT, Severity of Alopecia Tool; SALT50/90, at least 50/90% improvement from baseline in SALT score; SH, scalp hair; Wk, week.

✓Achieved statistical significance after adjustment for multiplicity

Did not achieve statistical significance after adjustment for multiplicity and stopped graphical testing scheme

✘p≤0.05, without adjustment for multiplicity

✘Not statistically significant, p>0.05

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 6. Results of graphical multiple-testing procedure in BRAVE-AA2

==> picture [454 x 393] intentionally omitted <==

Abbreviations : ClinRO, clinician-reported outcome; EB, eyebrow; EL, eyelash; PCFB, percent change from Baseline; PRO, patient-reported outcome; SALT, Severity of Alopecia Tool; SALT50/90, at least 50/90% improvement from baseline in SALT score; SH, scalp hair; Wk, week.

✓Achieved statistical significance after adjustment for multiplicity

Did not achieve statistical significance after adjustment for multiplicity and stopped graphical testing scheme ✘p≤0.05, without adjustment for multiplicity

✘Not statistically significant, p>0.05

B.2.4.5 Participant disposition

BRAVE-AA1

A total of 829 patients were screened in BRAVE-AA1, of whom 654 were randomised in a 2:2:3 ratio: 189 received placebo, 184 received baricitinib 2 mg, and 281 received baricitinib 4 mg.[67] Overall, study drug discontinuation rates ranged from 6.8% in the baricitinib 4 mg group to 11.1% in the placebo group. The most common reasons for discontinuation were withdrawal in placebo and baricitinib 4 mg groups (6.3% and 3.2%, respectively), and loss to follow-up in the baricitinib 2 mg group (3.3%). Patient disposition for the phase III portion of BRAVE-AA1 through Week 36 is shown in Figure 7.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 7. Patient disposition to Week 36 in BRAVE-AA1

==> picture [454 x 269] intentionally omitted <==

Source: BRAVE-AA1 Clinical Study Report.[65]

BRAVE-AA2

A total of 727 patients were screened in BRAVE-AA2, of whom 546 were randomised in a 2:2:3 ratio: 156 received placebo, 156 received baricitinib 2 mg, and 234 received baricitinib 4 mg.[67] Overall, study drug discontinuation rates ranged from 7.7% in the baricitinib 4 mg group to 13.5% in the placebo group. Across all groups, the most common reason for discontinuation was withdrawal by subject (placebo, 4.5%; baricitinib 2 mg, 3.8%; baricitinib 4 mg, 3.0%). Patient disposition for BRAVE-AA2 through Week 36 is shown in Figure 8.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 8. Patient disposition to Week 36 in BRAVE-AA2

==> picture [452 x 261] intentionally omitted <==

Footnotes:[ a ] For placebo, the other reasons included lack of adherence to study intervention and screening failure due to criteria 3 but randomised by mistake (this patient never received study intervention). b One patient interrupted study intervention prior to Week 36 and did not resume. Source: BRAVE-AA2 Clinical Study Report.[66]

B.2.5 Critical appraisal of the relevant clinical effectiveness evidence

The trials captured in the clinical SLR were assessed for quality using the Appraisal of RCT checklist by the Cochrane Collaboration,[69] while the quality of all observational studies was assessed using the quality assessment tool developed by the York University CRD.[70] The results of these quality assessments are presented in Appendix D, and a summary of the quality assessment for BRAVE-AA1 and -AA2 is presented in Table 15 below.

Table 14. Critical appraisal of BRAVE-AA1 and BRAVE-AA2

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Did the analysis include an intention-to-treat analysis? If so, was this appropriate and were appropriate methods used to account for missing Low Low data?

B.2.6 Clinical effectiveness results of the relevant studies

==> picture [452 x 355] intentionally omitted <==

----- Start of picture text -----
Summary of clinical effectiveness of baricitinib
• Treatment with baricitinib was associated with a statistically significant hair regrowth on the
scalp when compared to placebo. The proportion of patients achieving SALT≤20 at Week 36
in BRAVE-AA1 were 35.2%, 21.7%, and 5.3% for baricitinib 4 mg, 2 mg, or placebo,
respectively and 32.5%, 17.3%, and 2.6% in BRAVE-AA2
o Scalp hair regrowth following treatment with baricitinib 4 mg increased over 36 weeks
of treatment, and was associated with a statistically significant higher proportion of
patients achieving SALT≤20 at Week 16 and Week 24 as compared with placebo
o A higher proportion of patients receiving baricitinib 4 mg also achieved SALT≤10
(≥90% scalp hair coverage) compared with placebo at Week 36 in BRAVE-AA1 (26.0%
vs 3.7%; p<0.001), and in BRAVE-AA2 (23.5% vs 0.6%; p<0.001), representing almost
complete hair regrowth in these patients
• Treatment with baricitinib 4 mg led to a statistically significant eyelash and eyebrow regrowth
at Week 36 when compared with placebo treatment, with the proportion of patients achieving
an improvement in hair growth increasing over the treatment period
• The SALT≤20 response rate for baricitinib continued to increase from Week 36 through Week
52. Improvements in other secondary endpoints including SALT≤10, SALT50 and the ClinRO
Measures for EB or EL hair loss also continued to increase from Week 36 through Week 52.
• Among patients who achieved SALT≤20 at Week 52, *** of those who were randomised to
remain on baricitinib 4 mg in a down-titration sub-study, and ***** of those who were assigned
to remain on baricitinib 4 mg in a withdrawal sub-study achieved SALT≤20 at week 76,
suggesting that efficacy is maintained over time in most of the patients who have reached
SALT≤20. Among the patients who had reached SALT≤20 at week 52, the proportion who
had also achieved SALT≤10 increased up to week 76.
• Baricitinib was associated with a notable improvement in HRQoL when compared with
placebo. Statistically significant improvements were associated with baricitinib 4 mg treatment
after 36 weeks, as measured by the Skindex-16 AA tool across both trials and by HADS-
Depression and HADS-Anxiety scores in BRAVE-AA2
----- End of picture text -----

The anticipated license dose for baricitinib in AA is 4 mg once daily, although a dose of 2 mg once daily may be appropriate for some patients, or as a down-titration option if patients achieve sustained control of disease with 4 mg once daily. Baricitinib 4 mg will inform the base case. However, for completeness, data for both 4 mg and 2 mg treatment arms will be presented for both trials. Pooled data from BRAVE-AA1 and BRAVE-AA2 will also be presented in Section B.2.8 as these data will be used in the economic model.

B.2.6.1 Primary efficacy endpoint

Treatment with baricitinib is associated with significant improvement in hair regrowth on the scalp, eyebrow and eyelashes at Week 36

Proportion of patients achieving SALT≤20 at Week 36

The SALT score measures the extent of hair loss on the scalp, with higher scores representing a greater extent of hair loss on the scalp. SALT≤20, indicating less than 20% of scalp hair loss (or ≥80% scalp coverage with hair), represents a meaningful outcome for patients with ≥50% scalp hair loss.[71]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

In both BRAVE-AA1 and BRAVE-AA2, treatment with both baricitinib 4 mg and 2 mg resulted in a significantly higher proportion of patients achieving SALT≤20 at Week 36 compared to placebo, after adjusting for multiplicity (Table 15). In BRAVE-AA1, the proportions of patients achieving SALT≤20 at Week 36 were 35.2%, 21.7%, and 5.3% for baricitinib 4 mg, 2 mg, or placebo, respectively (Figure 9). In BRAVE-AA2, 32.5%, 17.3%, and 2.6% of patients who received baricitinib 4 mg, 2 mg, or placebo achieved SALT≤20 at Week 36 (p<0.001 for all comparisons with placebo in both trials) (Figure 10). Across both trials, only 5.3% and 2.6% of patients randomly assigned to placebo, respectively, achieved the primary endpoint of SALT≤20 at 36 weeks. These results confirm published observations that prognosis is particularly poor for patients with chronic extensive hair loss, and that the demonstration of efficacy in BRAVE-AA1 and BRAVE-AA2 was performed in a refractory population of patients. Results for the proportion of patients achieving SALT≤20 at Week 36 according to disease severity are presented in Appendix E.

Table 15. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Abbreviations: CI: confidence interval; PBO: placebo; SALT: Severity of Alopecia Areata Tool; NA = not applicable. Footnotes: Results in bold were statistically significant after adjustment for multiplicity.[a] Difference confidence intervals are constructed using Newcombe-Wilson method, without continuity correction. Response confidence intervals are constructed using Wilson method, without continuity correction.[b] Logistic regression analysis with treatment group, geographic region, duration of current episode at baseline (<4 years vs. ≥4 years), and baseline SALT score as factors. Source: BRAVE-AA1 Clinical Study Report. BRAVE-AA2 Clinical Study Report.[65, 66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 9. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI])

==> picture [454 x 254] intentionally omitted <==

Footnotes: ***p<0.001 baricitinib vs placebo . Abbreviations: AA: alopecia areata; BARI: baricitinib; NRI, non-responder imputation; PBO, placebo; SALT, Severity of Alopecia Tool. Source: BRAVE-AA1 Clinical Study Report.[65]

Figure 10. Proportion of patients achieving SALT≤20 at Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI])

==> picture [454 x 254] intentionally omitted <==

Footnotes: ***p<0.001 baricitinib vs placebo . Abbreviations: AA: alopecia areata; BARI: baricitinib; NRI, non-responder imputation; PBO, placebo; SALT, Severity of Alopecia Tool.

Source: BRAVE-AA2 Clinical Study Report.[66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.6.2 Secondary efficacy endpoints

An overview of results for key secondary endpoints (multiplicity adjusted) from BRAVE-AA1 and BRAVE-AA2 is provided in Table 16 and Table 17, respectively. Data regarding the most relevant endpoints are then described in more detail below.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 16. Efficacy results for key secondary endpoints (after adjustment for multiplicity) for BRAVE-AA1 (FAS population)

Footnotes: Results in bold denote statistical significance after adjustment for multiplicity; non-bold results denote significance but not after adjustment for multiplicity.[a] Denominator is number of patients for which data are available and is different from the total patient number in each treatment arm.[b ] Patients also achieved ≥2-point improvement from baseline.[c] Only assessed in patients with a score ≥3 at baseline.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved Page 56 of 151

Abbreviations : CI: confidence interval; ClinRO: clinician-reported outcome; eyebrow: eyebrow; eyelash: eyelash; NS: not significant; PRO: patient-reported outcome; SALT: Severity of Alopecia Tool; SALT50/90: ≥50/90% improvement from baseline in SALT score; SE: standard error. Source: BRAVE-AA1 Clinical Study Report; EPAR.[65, 67, 68]

Table 17. Efficacy results for key secondary endpoints (after adjustment for multiplicity) for BRAVE-AA2 (FAS population)

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes: Results in bold denote statistical significance after adjustment for multiplicity; non-bold results denote significance but not after adjustment for multiplicity.[a] Denominator is number of patients for which data are available and is different from the total patient number in each treatment arm.[b ] Patients also achieved ≥2-point improvement from baseline.[c] Only assessed in patients with a score ≥3 at baseline.

Abbreviations: CI: confidence interval; ClinRO: clinician-reported outcome; eyebrow: eyebrow; eyelash: eyelash; NS: not significant; PRO: patient-reported outcome; SALT: Severity of Alopecia Tool; SALT50/90: ≥50/90% improvement from baseline in SALT score; SE: standard error. Source: BRAVE-AA2 Clinical Study Report; EPAR.[66-68]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Treatment with baricitinib is associated with significant scalp hair regrowth in responders which increases over 36 weeks of treatment

Proportion of patients achieving SALT≤20 at Weeks 16 and 24

The proportion of patients achieving SALT≤20 at Weeks 16 and 24 across the phase III trials is summarised in Table 18. In BRAVE-AA1, a higher proportion of patients achieving SALT≤20 response was observed in patients treated with baricitinib 4 mg compared to placebo at both Week 16 (18.5% vs 4.2%; p≤0.001) and Week 24 (26.7% vs 4.8%; p≤0.001) after adjustment for multiplicity. A statistically significant increase in the proportion of patients achieving SALT≤20 at Week 24 was also observed in the baricitinib 2 mg group compared to placebo (p≤0.05). In BRAVE-AA2, the proportion of patients receiving baricitinib 4 mg who achieved SALT≤20 was similarly higher compared with placebo at Week 24 (28.2% vs 1.3%; p≤0.001) after adjusting for multiplicity. Statistically significant increases in the proportion of patients achieving SALT≤20 at Week 16 was also observed in both 2 mg (p≤0.01) and 4 mg treatment groups (p≤0.001) (not adjusted for multiplicity).

Overall, the SALT≤20 response rates for baricitinib in each trial continued to increase over the 36 weeks of treatment (Figure 11, Figure 12 and Figure 13). Results for the response rates over time through Week 36 according to disease severity are presented in Appendix E.

Table 18. Proportion of Patients achieving SALT≤20 at Weeks 16 and 24 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes : Results in bold were statistically significant after adjustment for multiplicity.[a] Difference confidence intervals are constructed using Newcombe-Wilson method, without continuity correction. Response confidence intervals are constructed using Wilson method, without continuity correction.[b] Logistic regression analysis with treatment group, geographic region, duration of current episode at baseline (<4 years vs. ≥4 years), and baseline SALT score as factors.

Abbreviations: SALT: Severity of Alopecia Areata Tool; NA: not applicable. Source: BRAVE-AA1 Clinical Study Report. BRAVE-AA2 Clinical Study Report.[65-67]

Figure 11. Proportion of patients achieving SALT≤20 over time through Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI])

==> picture [454 x 229] intentionally omitted <==

Footnotes: Statistically significant after adjustment for multiplicity (only Weeks 16, 24, and 36 were included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p<0.001 vs placebo. Abbreviations: AA: alopecia areata; BARI: baricitinib; FAS: full analysis set; NRI: non-responder imputation; PBO: placebo; SALT: Severity of Alopecia Tool. Source: BRAVE-AA1 Clinical Study Report; EPAR.[65, 68]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 12. Proportion of patients achieving SALT≤20 over time through Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI])

==> picture [454 x 246] intentionally omitted <==

Footnotes: Statistically significant after adjustment for multiplicity (only Weeks 16, 24, and 36 were included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p<0.001 vs placebo. Abbreviations: AA: alopecia areata; BARI: baricitinib; FAS: full analysis set; NRI: non-responder imputation; PBO: placebo; SALT: Severity of Alopecia Tool. Source: BRAVE-AA2 Clinical Study Report; EPAR.[66, 68]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 13. Photographs of changes in absolute SALT scores over time through Week 36 in patients treated with baricitinib in BRAVE-AA1

==> picture [452 x 451] intentionally omitted <==

Abbreviations : BARI: baricitinib; SALT: Severity of Alopecia Tool.

Proportion of patients achieving SALT50 at Week 12

Across both trials, both 2 mg and 4 mg baricitinib demonstrated meaningful improvements in scalp hair regrowth before Week 36, with a statically significant increase in the proportion of patients achieving SALT50 at Week 12 compared with placebo. In BRAVE-AA1, 21.7% of patients in the baricitinib 4 mg treatment group achieved SALT50 at Week 12 compared with only 4.8% of patients in the placebo group (p<0.001). Significantly more patients in the 2 mg group also achieved this endpoint compared with placebo at Week 12 (p<0.05). Similar results were observed in BRAVE-AA2 with significantly more patients treated with baricitinib 4 mg or 2 mg having SALT50 at Week 12 versus placebo (23.5% and 10.9% vs 2.6%; p<0.001 [4 mg] and p<0.01 [2 mg]).

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 19. Proportion of Patients Achieving a SALT50 at Week 12 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Footnotes:[a] Difference confidence intervals are constructed using Newcombe-Wilson method, without continuity correction. Response confidence intervals are constructed using Wilson method, without continuity correction. b Logistic regression analysis with treatment group, geographic region, duration of current episode at baseline (<4 years vs. ≥4 years), and baseline SALT score as factors. Abbreviations: SALT50: at least 50% improvement from Baseline in Severity of Alopecia Areata Tool score; NA: not applicable. Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Proportion of patients achieving SALT50 at Week 36

Across both trials, both 2 mg and 4 mg baricitinib demonstrated meaningful improvements in scalp hair regrowth from baseline at Week 36, with a statically significant increase in the proportion of patients achieving SALT50 (at least 50% improvement from baseline) at Week 36 compared with placebo. The proportion of patients achieving SALT50 at Week 36 across the phase III trials is summarised in Table 20. In BRAVE-AA1, ***** and ***** of patients in the baricitinib 4 mg and 2 mg groups, respectively, achieved SALT50 at Week 36, compared with ***** of placebo-treated patients (*******; not adjusted for multiplicity). In BRAVE-AA2, ***** and ***** of patients treated with baricitinib 4 mg or 2 mg, respectively, achieved SALT50 at Week 36 versus **** of those treated with placebo (*** **** not adjusted for multiplicity).

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 20. Proportion of patients achieving an absolute SALT50 at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Footnotes:[a] Logistic regression analysis with treatment group, geographic region, duration of current episode at Baseline (<4 years vs. ≥4 years), and baseline SALT score as factors. Abbreviations: SALT50: at least 50% improvement from Baseline in Severity of Alopecia Areata Tool score; PBO: placebo; NA: not applicable. Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Proportion of patients achieving SALT75 at Week 36

Baricitinib 2 mg and 4 mg also demonstrated a statically significant increase in the proportion of patients achieving SALT75 (at least 75% improvement from baseline) at Week 36 compared with placebo. The proportion of patients achieving SALT75 at Week 36 across the phase III trials is summarised in Table 21. In BRAVE-AA1, ***** and ***** of patients in the baricitinib 4 mg and 2 **** mg groups, respectively, achieved SALT75 at Week 36, compared with of placebo-treated patients (*******; not adjusted for multiplicity). In BRAVE-AA2, ***** and ***** of patients treated with baricitinib 4 mg or 2 mg, respectively, achieved SALT75 at week 36 versus **** of those treated with placebo (*** **** not adjusted for multiplicity).

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 21. Proportion of patients achieving an absolute SALT75 at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Footnotes:[a] Logistic regression analysis with treatment group, geographic region, duration of current episode at Baseline (<4 years vs. ≥4 years), and baseline SALT score as factors. Abbreviations: SALT50: at least 50% improvement from Baseline in Severity of Alopecia Areata Tool score; NA: not applicable.

Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Treatment with baricitinib is associated with almost complete hair regrowth on the scalp by Week 36 in a subset of patients

Proportion of patients achieving an absolute SALT≤10 at Weeks 24 and 36

The proportion of patients achieving SALT≤10 (≥90% scalp coverage with hair) at Weeks 24 and 36 across the phase III trials is summarised in Table 22. Compared with placebo, a higher proportion of patients randomised to receive baricitinib 4 mg achieved SALT≤10 at Weeks 24 (***** ** ***** *******) and 36 (***** ** ***** *******) in BRAVE-AA1, and at Week 36 in BRAVE-AA2 (***** ** ***** *******) after adjustment for multiplicity. In BRAVE-AA1, there was also a higher proportion of patients in the baricitinib 2 mg group that achieved SALT≤10 at Weeks 24 (**** ** ***** ******) and 36 (***** ** ***** ******) versus placebo after adjustment for multiplicity (Figure 14).

Consistent with SALT ≤20, the SALT≤10 response rates for both doses of baricitinib continued to increase over the 36 weeks of treatment (Figure 14 and Figure 15). In addition, most patients treated with baricitinib 4 mg who had reached SALT≤20 by Week 36 also achieved SALT≤10.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 22. Proportion of Patients Achieving an Absolute SALT≤10 at Weeks 24 and 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

Abbreviations: SALT≤10: less than 10% of hair loss observed using the Severity of Alopecia Areata Tool score; NA: not applicable.

Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 14. Proportion of patients achieving SALT≤10 through Week 36 in BRAVE-AA1 and BRAVE-AA1 (FAS population; primary censoring rule [NRI])

==> picture [454 x 232] intentionally omitted <==

Footnotes:[a ] Statistically significant after adjustment for multiplicity (only Weeks 24 and 36 were included in the graphical testing procedure). *p<0.05 vs placebo; **p<0.01 vs placebo; ***p<0.001 vs placebo. Abbreviations: AA, alopecia areata; BARI, baricitinib; FAS, full analysis set; PBO: placebo; SALT: Severity of Alopecia Tool.

Source: BRAVE-AA1 Clinical Study Report.[65]

Figure 15. Proportion of patients achieving SALT≤10 through Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [NRI])

==> picture [454 x 226] intentionally omitted <==

Footnotes:[a ] Statistically significant after adjustment for multiplicity (only Weeks 24 and 36 were included in the graphical testing procedure). *p<0.05 vs placebo; **p<0.01 vs placebo; ***p<0.001 vs placebo. Abbreviations: AA, alopecia areata; BARI, baricitinib; FAS, full analysis set; PBO: placebo; SALT: Severity of Alopecia Tool.

Source: BRAVE-AA2 Clinical Study Report.[66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Treatment with baricitinib is associated with significant eyebrow and eyelash regrowth after 36 weeks

Proportion of patients achieving ClinRO measure for eyebrow hair loss 0 or 1 at Week 36

The ClinRO measure for eyebrow hair loss assesses the extent of eyebrow loss, with higher scores representing greater hair loss. The proportion of patients with a score ≥2 at baseline achieving ClinRO measure for eyebrow hair loss 0 or 1 at Week 36 is summarised in Figure 16 and Figure 17. A higher proportion of patients treated with baricitinib 4 mg achieved a score of 0 or 1 with a ≥2-point improvement from baseline in the ClinRO Measure for eyebrow Hair Loss at Week 36 compared with placebo after adjustment for multiplicity in both BRAVE-AA1(31.4% vs 3.2%; p<0.001) and BRAVE-AA2 (34.8% vs. 4.5%; p<0.001). Baricitinib 2 mg also demonstrated a statistically significant improvement compared with placebo in the proportion of patients achieving ClinRO Measure for eyebrow Hair Loss score of 0 or 1 at Week 36 in BRAVE-AA1 (3.2% vs 19.1%; p<0.001). The proportion of patients treated with baricitinib who achieved a score of 0 or 1, with a ≥2-point improvement from baseline in the ClinRO Measure for eyebrow Hair Loss continued to increase through Week 36.

Figure 16. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyebrow Hair Loss™ through Week 36 in BRAVE-AA1 (FAS population; primary censoring rule [NRI])

==> picture [454 x 247] intentionally omitted <==

Footnotes: Patients with score of ≥2 at baseline.[a] Statistically significant after adjustment for multiplicity (only Week 36 was included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p≤0.001 vs placebo.

Abbreviations: BARI: baricitinib; ClinRO: clinician-reported outcome; eyebrow: eyebrow; FAS: full analysis set; PBO: placebo. Source: BRAVE-AA1 Clinical Study Report.[65]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 17. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyebrow Hair Loss™ through Week 36 in BRAVE-AA2 (FAS population; primary censoring rule [NRI])

==> picture [454 x 244] intentionally omitted <==

Footnotes: Patients with score of ≥2 at baseline.[ a] Statistically significant after adjustment for multiplicity (only week 36 was included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p≤0.001 vs placebo.

Abbreviations: BARI: baricitinib; ClinRO: clinician-reported outcome; eyebrow: eyebrow; FAS: full analysis set; PBO: placebo. Source: BRAVE-AA2 Clinical Study Report.[66]

Proportion of patients achieving ClinRO measure for eyelash hair loss 0 or 1 at Week 36

The ClinRO measure for eyelash hair loss assesses the extent of eyelash hair loss, with higher scores representing greater eyelash hair loss The proportion of patients with score of ≥2 at baseline achieving ClinRO measure for eyelash hair loss 0 or 1 at Week 36 is summarised in Figure 18 and Figure 19. A higher proportion of patients treated with baricitinib 4 mg achieved a score of 0 or 1 with a ≥2-point improvement from baseline in the ClinRO Measure for eyelash Hair Loss at week 36 compared with placebo after adjustment for multiplicity in in BRAVE-AA1 (33.5% vs 3.1%; p<0.001) and BRAVE-AA2 (34.3% vs 5.6%; p<0.001). Baricitinib 2 mg did not achieve a statistically significant improvement compared to placebo in either trial. Nevertheless, similar findings were observed with respect to improvements over time, as the proportion of patients treated with baricitinib who achieved a score of 0 or 1 for eyelash Hair Loss continued to increase through to Week 36.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Figure 18. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyelash Hair Loss™ through Week 36 in BRAVEAA1 (FAS population; primary censoring rule [NRI])

==> picture [454 x 232] intentionally omitted <==

Footnotes: Patients with a score of ≥2 at baseline[a] Statistically significant after adjustment for multiplicity (only Week 36 was included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p≤0.001 vs placebo.

Abbreviations: BARI: baricitinib; ClinRO: clinician-reported outcome; eyelash: eyelash; FAS: full analysis set; PBO: placebo.

Source: BRAVE-AA1 Clinical Study Report.[65]

Figure 19. Proportion of patients achieving a score of 0 or 1 with ≥2-point improvement from baseline on the ClinRO Measure for eyelash Hair Loss™ through Week 36 in BRAVEAA2 (FAS population; primary censoring rule [NRI])

==> picture [454 x 229] intentionally omitted <==

Footnotes: Patients with a score of ≥2 at baseline.[a] Statistically significant after adjustment for multiplicity (only Week 36 was included in the graphical testing procedure).*p<0.05 vs placebo; **p<0.01 vs placebo; ***p≤0.001 vs placebo. Abbreviations: BARI: baricitinib; ClinRO: clinician-reported outcome; eyelash: eyelash; FAS: full analysis set; PBO: placebo.

Source: BRAVE-AA2 Clinical Study Report.[66]

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

B.2.6.3 Health-related quality of life (HRQoL) outcomes

Treatment with baricitinib demonstrates a notable improvement in HRQoL when compared with placebo

Mean change from baseline in Skindex-16 AA domain scores at Week 36

The Skindex-16 adapted for AA score assesses the impact of AA on quality of life by measuring the degree to which the subjects are bothered by AA and associated symptoms. The Skindex-16 scores in patients enrolled in BRAVE-AA1 and BRAVE-AA2 across the 3 domains (emotions, functioning and symptoms) are shown in Table 23. Overall, baricitinib 4 mg and 2 mg demonstrated a statistically significant improvement in mean change from baseline to Week 36 in the Skindex-16 AA Emotions domain compared with placebo in both the BRAVE-AA1 and BRAVE-AA2 trials (*******). In addition, baricitinib 4 mg was associated with a statistically significantly greater improvement in the Functioning domain compared with placebo at Week 36 in both phase III trials (BRAVE-AA1: ******; BRAVE-AA2: *******). Finally, statistically significant differences in the Symptoms domain versus placebo were observed in patients treated with ****** ****** baricitinib 4 mg in BRAVE-AA2 ( ) and 2 mg in BRAVE-AA1 ( ).

Table 23. Mean change from baseline in Skindex-16 AA domain scores (Emotions, Functioning and Symptoms) at Week 36 in BRAVE-AA1 and BRAVE-AA2 (FAS population; primary censoring rule [mLOCF])

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Footnotes: Baseline is defined as the last non-missing assessment recorded on or prior to the date of first study drug administration. If a patient is randomised but does not receive study drug, then the date of randomization is used instead of the first dose date.[a] number of BRAVE-AA1 patients: n=110; baricitinib 2 mg, n=102; baricitinib 4 mg, n=165; number of BRAVE-AA2 patients: placebo, n=146; baricitinib 2 mg, n=147; baricitinib 4 mg, n=227. Abbreviations: AA: alopecia areata; NA: not applicable; LSM, least squares mean; Skindex-16 AA: Skindex-16 Adapted for Alopecia areata; SD: standard deviation; SE: standard error. Source: BRAVE-AA1 Clinical Study Report; BRAVE-AA2 Clinical Study Report.[65, 66]

Skindex-16 AA domain scores at Week 36: Responder analyses

A separate analysis of the Skindex-16 AA scores by SALT50 responder status at Week 36 (responders versus non-responders) was conducted. The Skindex-16 scores in patients enrolled in the BRAVE-AA studies (pooled Week 36 efficacy population) across the 3 domains (emotions, functioning and symptoms) for responders and non-responders are shown in Table 24.

Overall, baricitinib 4 mg and 2 mg demonstrated a statistically significant improvement in mean change from baseline for responders versus non-responders in the Emotions domain at Week 36 in the pooled efficacy analysis (). Statistically significant improvements in mean change for responders versus non-responders for baricitinib 4 mg and 2 mg were also observed for the Functioning domain ( and *******, respectively). For the Symptoms domain, a numerical improvement in mean change for responders versus non-responders was observed, however this difference was not statistically significant. These data demonstrate that achieving a SALT50 response at Week 36 is associated with notable improvement in Skindex-AA scores, particularly in the Emotions and Functions domains that are likely most affected by AA.

Company evidence submission template for baricitinib for treating severe alopecia areata [ID3979]

© Eli Lilly (2022). All rights reserved

Table 24: Mean change in Skindex-16 AA domain scores (Emotions, Functioning and Symptoms) at Week 36 in the BRAVE-AA studies for SALT50 Responders versus Non-responders (pooled Week 36 efficacy population; primary censoring rule [mLOCF])