Single Technology Appraisal

Voclosporin with immunosuppressive therapies for treating lupus nephritis [ID3962]

Committee Papers

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

SINGLE TECHNOLOGY APPRAISAL

Voclosporin with immunosuppressive therapies for treating lupus nephritis [ID3962]

Contents:

The following documents are made available to stakeholders:

Access the final scope and final stakeholder list on the NICE website.

1. Company submission from Otsuka Pharmaceuticals

2. Clarification questions and company responses

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

• a. AOFAC Foundation

• b. Lupus UK

• c. UK Kidney Association

• d. UK Renal Pharmacy Group

4. External Assessment Report prepared by Peninsula Technology Assessment Group (PenTAG)

5. External Assessment Report – factual accuracy check

6. Technical engagement response from company

7. Technical engagement responses and statements from experts:

• a. Sian Brennan – patient expert, nominated by LUPUS UK

• b. Professor Alan Salama, Professor of Nephrology – clinical expert, nominated by LUPUS UK

• c. Amy Somers – patient expert nominated by LUPUS UK

8. Technical engagement responses from consultees and commentators:

• a. British Society of Rheumatology – endorsed by Royal College of Physicians

• b. NHS England

• c. NHS England – Specialised rheumatology clinical reference group d. Novartis

9. External Assessment Group critique of company response to technical engagement prepared by Peninsula Technology Assessment Group (PenTAG)

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

Voclosporin with immunosuppressive therapies for treating lupus nephritis [ID3962]

Document B

Company evidence submission

May 2022

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Contents

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

List of tables

==> picture [452 x 706] intentionally omitted <==

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

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

==> picture [451 x 733] intentionally omitted <==

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

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

Table B.3-19. Grade III/IV TEAE costs and frequencies .................................................... 139 Table B.3-20. Key model assumptions .............................................................................. 143 Table B.3-21. Base-case results (discounted) ................................................................... 145 Table B.3-22. Base-case results (undiscounted) ............................................................... 145 Table B.3-23. Net health benefit ........................................................................................ 146 Table B.3-24. Mean results of PSA (1000 simulations) and comparison with base-case results ............................................................................................................................... 146 Table B.3-25. Scenario analyses (voclosporin + MMF versus MMF) ................................. 149 Table B.3-26. Scenario analysis results following adjustment of treatment duration from 36 months to 18 months (apart from TAC-containing regimens) ............................................ 150 Table B.3-27. Mean results of PSA (1000 simulations) for scenario and comparison with deterministic scenario results ............................................................................................ 150 Table B.3-28. Internal validation of model outputs at 12 months ....................................... 152

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 6 of 165

List of figures

Figure B.1-1. Voclosporin mechanism of action .................................................................. 12 Figure B.2-1. Trial design: AURORA 1 (AUR-VCS-2016-01; NCT03021499) ...................... 29 Figure B.2-2. Trial design: AURORA 2 ................................................................................ 40 Figure B.2-3. AURA-LV: Trial Design (AUR-VCS-2012-01; NCT02141672) ........................ 48 Figure B.2-4. AURORA 1: Probability of UPCR of ≤0.5 mg/mg ........................................... 63 Figure B.2-5. AURORA 1: Probability of ≥ 50% Reduction from Baseline in UPCR ............. 64 Figure B.2-6: AURA-LV: Analysis of Time to CRR .............................................................. 71 Figure B.2-7. Treatment network for studies contributing to evidence for CRR in the overall population ........................................................................................................................... 76 Figure B.2-8. Treatment network for studies contributing to evidence for PRR in the overall population ........................................................................................................................... 77 Figure B.2-9. Forest plot for posterior median ORs and 95% CrI, for CRR .......................... 80 Figure B.2-10. Forest plot for posterior median ORs and 95% CrI, for PRR ........................ 83 Figure B.3-1. Cost-effectiveness analysis model structure ................................................ 112 Figure B.3-2. Time to study drug discontinuation (AURORA 1 and AURORA 2) ............... 121 Figure B.3-3. TTD log-cumulative hazard plot (AURORA 1 and AURORA 2) .................... 122 Figure B.3-4. TTD Schoenfeld residual plot (AURORA 1 and AURORA 2) ....................... 123 Figure B.3-5. Scatter plot of PSA results for total discounted costs and QALYs ................ 147 Figure B.3-6. Scatter plot of PSA results for incremental discounted costs and QALYs (voclosporin + MMF vs comparators) ................................................................................ 147 Figure B.3-7. Cost-effectiveness acceptability curve ......................................................... 147 Figure B.3-8. DSA tornado diagram - incremental costs for voclosporin + MMF vs MMF .. 148 Figure B.3-9. DSA tornado diagram - incremental QALYs for voclosporin + MMF vs MMF 148 Figure B.3-10. DSA tornado diagram – ICER (£/QALY) for voclosporin + MMF vs MMF* . 149 Figure B.3-11. Scenario – Scatter plot of PSA results for total discounted costs and QALYs ......................................................................................................................................... 151 Figure B.3-12. Scenario - Scatter plot of PSA results for incremental discounted costs and QALYs (voclosporin + MMF vs comparators) .................................................................... 151 Figure B.3-13. Scenario – Cost-effectiveness acceptability curve ..................................... 151

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 7 of 165

Abbreviations

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.1. Decision problem, description of the technology and clinical care pathway

B.1.1 Decision problem

The submission covers the technology’s full marketing authorisation for the following indication:

• Voclosporin is indicated in combination with background immunosuppressive therapies for the treatment of adult patients with active class III, IV or V (including mixed class III/V and IV/V) lupus nephritis (LN)

A summary of the decision problem is provided in Table B.1-1.

Table B.1-1. The decision problem

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.1.2 Description of the technology being evaluated

Voclosporin (Lupkynis™) is a novel orally administered next generation calcineurin inhibitor (CNI) immunosuppressant with a dual mechanism of action which reduces proinflammatory T-cell mediated immune responses linked to kidney inflammation,[1] and protects renal podocytes from damage (Figure B.1-1).[2]

Specifically, voclosporin binds to calcineurin and blocks calcineurin-mediated activation of Nuclear Factor of Activated T-Cells (NFAT), a transcription factor which drives T-cell immune response.[1,3-6] CNI immunosuppressive activity results in inhibition of lymphocyte proliferation, T-cell cytokine production, and expression of T-cell activation surface antigens.[1] In addition, studies in animal models indicate that voclosporin stabilises actin cytoskeleton and stress fibres in renal podocytes, leading to increased podocyte integrity in glomeruli.[1] Podocytes are specialised epithelial cells that are a key component of the glomerular filtration barrier, and their cytoskeletal integrity is critical to ensure healthy kidney function.[4-7]

Figure B.1-1. Voclosporin mechanism of action

==> picture [452 x 272] intentionally omitted <==

Abbreviations: APC = antigen-presenting cell; IL = interleukin; LN = lupus nephritis; TNF = tumour necrosis factor

Voclosporin’s novel molecular structure and mechanism of action may eliminate the need for regular therapeutic drug monitoring required with currently available CNIs (ciclosporin and tacrolimus), and potentially minimise the risk of CNI-associated side effects such as diabetes, kidney dysfunction, and hypertension (Section B.1.3.7 and Section B.2.10). Voclosporin is structurally similar to ciclosporin, but incorporates a modification to a functional group on amino acid-1 of the molecule.[8] This modification changes both how voclosporin binds to calcineurin and its metabolic profile, leading to a four-fold increase in immunosuppressive potency compared to ciclosporin and fewer CNI-associated side effects due to the rapid elimination of voclosporin metabolites.[8] In addition, the combination of increased potency and decreased metabolite exposure gives voclosporin a more predictable pharmacokinetic and

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

pharmacodynamic profile compared to currently used CNIs, eliminating the need for intensive therapeutic monitoring.[8-11]

A summary of the technology being appraised, voclosporin, is provided in Table B.1-2.

Table B.1-2. Technology being evaluated

Abbreviations: BID: twice daily; CHMP = Committee for Medicinal Products for Human Use; CNI = calcineurin inhibitors; EMA = European Medicines Agency; IL = interleukin; LN = lupus nephritis; MAA = marketing authorisation application; MMF = mycophenolate mofetil; MHRA = Medicines and Healthcare products Regulatory Agency; NFAT = nuclear factor of activated T-Cells; PAS = Patient Access Scheme; SmPC = summary of product characteristics; UK = United Kingdom

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

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

B.1.3.1 Disease overview

Systemic lupus erythematosus (SLE) is a chronic and complex autoimmune disease that can affect any organ in the body.[12] In SLE, abnormal and persistent immune system reactions to autologous nucleic acids result in the formation of damaging deposits of immune cell and autologous cellular materials called immune complex deposits.[13,14] These immune complexes form within organ systems throughout the body (e.g. skin, joints, kidney, and central nervous system).[13]

Lupus nephritis (LN) is the most common serious manifestation of SLE, affecting at least a third of patients,[15] although this may be as high as 60% among those with Black or Hispanic family backgrounds.[16-18] LN is characterised by the formation of immune complex deposits within renal tissues, leading to inflammation of the kidneys, renal damage, proteinuria and impaired renal function.[12,14]

LN-associated renal inflammation and structural/functional damage to renal cells is caused by the production of local cytokines, chemokines and adhesion molecules, along with an ensuing influx of inflammatory cells and proinflammatory cytokines.[14] T-cells play a major role in the pathogenesis and progression of LN, and contribute to renal tissue injury both directly and indirectly.[19-23] T-cells amplify inflammation by producing inflammatory cytokines, and also cause renal cell damage either by direct cytotoxicity, or through activation of macrophages, natural killer cells, dendritic cells and/or nephritogenic auto-antibody producing B cells.[23-25] LN is also associated with the disruption of podocyte function. Podocytes are highly specialised epithelial cells which form part of the filtration barrier in the kidneys, and are important in the regulation of glomerular filtration and regulation of protein loss.[26]

LN is an incurable, debilitating and potentially life-threatening disease that can cause permanent kidney damage.[12,27] If LN is left untreated, patients will progress through the stages of chronic kidney disease (CKD1-5), and may even go on to develop end-stage renal disease (ESRD) i.e. CKD5.[12,27] Overall, ESRD develops in 10–30% of patients with LN.[18,28] ESRD has particularly severe clinical consequences for patients, including high mortality rates and the need for invasive kidney replacement therapy, such as dialysis and/or kidney transplantation.[27]

B.1.3.2 Epidemiology

SLE is estimated to be prevalent in around 60,000 people in England and Wales and there are around 3,000 new SLE diagnoses each year (based on data collected within a retrospective cohort between 1999–2012 [LN data not available]).[29,30] Between 7–31% of patients with SLE have LN at diagnosis, and many go on to develop LN during the course of SLE disease (~30% within 1 year, ~40% within 5 years, and 40–48% within 15 years).[18] Guidelines from the British Society for Rheumatology (BSR) state that about one-third of SLE patients in the United Kingdom (UK) develop LN.[15]

Data describing the prevalence and incidence of LN in the UK are currently limited. Among publicly available data, the most recent UK-specific study was a 2001 retrospective analysis conducted in England, which reported overall LN prevalence and incidence rates of 4.4 and 0.4 per 100,000 of the population, respectively.[31]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 14 of 165

Prevalence and incidence rates are also known to be higher among certain subsets of the UK population. According to the same English study, approximately 85% of LN cases were in women (female vs male prevalence: 7.1 and 1.4 per 100,000; female vs male incidence: 0.7 vs 0.1 per 100,000, respectively [2001]) and most developed the disease when they were of childbearing age (age 18–39: 7.7 per 100,000; 40–59: 9.6 per 100,000; ≥60: 3.5 per 100,000).[31,32] LN was also more prevalent in Indo-Asian (12.6 per 100,000), African-Caribbean (60.8 per 100,000) and Chinese (65.5 per 100,000) populations compared to those of White ethnicity (3.5 per 100,000).[31]

In the absence of more recent published epidemiology data in England and Wales; overall LN incidence and prevalence can instead be estimated by considering the proportion of the total population who have SLE, the proportion of patients with SLE diagnosed with LN, and also the proportion of patients with SLE and LN who have class III–V active LN specifically. Based on a total population of 59,719,724 people in England and Wales (2020), this would equate to a prevalence of 13,521 patients with active class III, IV or V LN and an incidence of 684 new diagnoses per year. A summary of this calculation is presented in Table B.1-3.

Table B.1-3. LN epidemiology estimates for England and Wales in 2020

Abbreviations: LN = lupus nephritis; ONS = Office for National Statistics Source: Office for National statistics 2021[30]

B.1.3.3 Symptomatology and clinical presentation

Clinical presentation of LN is often subtle, and most commonly revealed by examination of the urine and blood.[12] Proteinuria is the defining aspect of LN and indicates both disease activity and kidney damage. Therefore, once proteinuria is clinically apparent, kidney tissues are already inflamed and damaged.[12,27] The most common clinical signs of LN (and approximate

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

prevalence) include proteinuria (100%), microscopic haematuria (80%), renal insufficiency (60%), nephrotic syndrome (50%), red blood cells (30%) or other cellular casts in urine (30%), and hypertension (30%).[12] Although patients with LN may experience few or no accompanying symptoms, a substantial proportion of patients may also experience skin rash across the nose and cheeks (~31%), photosensitivity (~8%), oral ulcer (~12%), arthritis (~6%), serositis (~24%), neurologic disorder (~1%), hematologic disorder (~89%), and/or immunologic disorder (~93%).[33-35]

The overarching goal of LN treatment is to quickly reduce proteinuria and inflammation to prevent further kidney damage.[12,27] However, renal flares occur in approximately 27–66% of LN patients,[36] usually within 5 to 6 years following the start of treatment.[27] The European Alliance of Associations for Rheumatology (previously European League Against Rheumatism) and European Renal Association–European Dialysis and Transplant Association (EULAR/ERA-EDTA) define a renal flare as an increase in proteinuria or serum creatinine level, an abnormal urinary sediment, or a reduction in creatinine clearance due to active disease.[15] Renal flares can be subdivided into proteinuric or nephritic flares:[36]

• Proteinuric flares – persistently increased proteinuria (>0.5–1.0 g daily) after a complete response (CR), or doubling of proteinuria (to >1.0 g daily) after a partial response (PR)

• Nephritic flares – an increase or recurrence of urinary sediment with or without increased proteinuria and are usually associated with a decline in renal function

Thus, renal flares result in histological progression to more severe disease (i.e. further kidney damage and decreased renal function) in 40–76% of patients, with rates of progression varying according to LN class (see section B.1.3.4).[18,27,37]

B.1.3.4 Disease classifications

LN severity is classified into LN class I to VI, by kidney biopsy according to the International Society of Nephrology/Renal Pathology Society (ISN/RPS) 2003 classification system (summarised in Table B.1-4). In some cases, biopsies may show mixed histological findings, warranting a combination of classifications (e.g. classes III + V, or class IV + V).[38] At initial LN diagnosis, the majority of patients are diagnosed with class III (10–25%), class IV (35–60%), and class V (5–30%) disease; while fewer patients are diagnosed at classes I, II, and VI (class I: 0–6%; class II: 1–20%; class VI: <5%).[18] Treatment decisions are largely based on the type and extent of renal damage.[28,39] For example, patients in classes I and II generally do not require treatment, while those in classes III, IV, and V benefit from potent immunosuppression and patients in class VI are considered for renal replacement therapy.[27]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 16 of 165

Table B.1-4. Summary of ISN/RPS classification of LN

Abbreviations: IC = immune complex; ISN/RPS = International Society of Nephrology/Renal Pathology Society; LN = lupus nephritis Source: Weening 2004[40]

Furthermore, patients may be classified according to their level of renal function (i.e. estimated glomerular filtration rate [eGFR]). If disease remains uncontrolled, patients with LN will progress through the stages of CKD (CKD1: >90 ml/min/1.73m[2] ; CKD2: 60–89 ml/min/1.73m[2] ; CKD3: 30–59 ml/min/1.73m[2] ; CKD4: 15–29 ml/min/1.73m[2] ) to ESRD (CKD5: <15 ml/min/1.73m[2] ).[12,41-43]

B.1.3.5 Burden to patients, carers and society

==> picture [43 x 9] intentionally omitted <==

Clinical burden

B.1.3.5.1.1 Disease progression and mortality risk

Progressive, uncontrolled kidney damage drives the clinical burden of LN.[44-46] Despite treatment, patients remain at high risk of renal flares, which may cause further renal damage and increase the likelihood of progression to CKD and ESRD.[36,47] In England, a retrospective analysis indicates that around 8% of patients with LN develop ESRD within 5 years of diagnosis (n=86; 1996–2005); while up to 20% of patients develop ESRD within three decades (n=154; 1975–2005).[48] More recently, studies outside of the UK (including a comprehensive literature review and meta-analysis) have reported even higher rates of progression to ESRD for patients with LN (10–50%).[16,18,28,49,50]

LN is associated with considerable mortality risk; however, progression to ESRD has particularly severe clinical consequences, including higher mortality rates and the need for invasive kidney replacement therapy (i.e. dialysis and/or kidney transplantation).[27] Although there are limited mortality data for LN in the UK, studies outside the UK associate LN with a 6–9-fold increase in mortality risk relative to a general population, which increases to a 26fold-greater risk if the disease progresses to ESRD.[28,44,45] Similarly, a multi-national cohort study which included patients from England and Wales suggests that that LN is significantly more lethal than SLE alone (hazard ratio [HR] = 2.98 [95% CI: 1.48, 5.99]; p=0.002; n=1,827).[16,51] In England, five-year mortality rates increase substantially from 4.7% for patients Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

with LN (1996–2005 [n=86])[48] to 36.5% of patients with ESRD (2003–2005 [n=750]).[52] Although dialysis and kidney transplantation are effective in reducing mortality among patients with ESRD, most patients receive dialysis in a clinic which requires a 4–8-hour procedure at least 3 times per week until a kidney donor becomes available (2.5–3 years average waiting time).[27,53,54] In some cases, patients may receive dialysis at home.[55] However, these patients would still be limited to the confines of their home for extended periods of time, with duration and intensity dependent on the patient’s needs. The National Health Service (NHS) recommend a variety of home dialysis schedules such as four days a week for four hours; five days a week for three hours; and six days a week for eight hours overnight.[55]

B.1.3.5.1.2 Pregnancy

As well as disease progression and mortality, LN is linked with poor maternal and foetal outcomes.[56] This is particularly important, given that the majority of patients with LN are women (~85%) (Section B.1.3.2) and most develop the disease when they are of childbearing age.[32,57,58] LN at the time of conception, or a history of prior LN, are both significantly associated with maternal hypertension (p<0.001), while prior LN is associated with an increased risk for preeclampsia.[59] High rates of preterm birth (39.4%), intrauterine growth restriction (12.7%), stillbirth (3.6%), and neonatal death (2.5%) have also been reported among LN-associated pregnancies.[59,60] LN-related kidney impairment may even cause infertility due to hypothalamic–pituitary dysfunction, and manifest as menstrual irregularity (including anovulatory cycles) in women or erectile dysfunction with reduced spermatogenesis in men.[61] Disease-related pregnancy concerns are further exacerbated by the use of treatments which may impair fertility and/or be harmful to a foetus.[62,63]

==> picture [44 x 9] intentionally omitted <==

Humanistic burden

Although there are limited UK-specific data, LN is generally associated with poor healthrelated quality of life (HRQoL)[64-67] due to both the symptomatic burden of LN (Section B.1.3.3)[33] and adverse effects associated with treatments (see Section B.1.3.7).[64] Patients with LN have reported HRQoL impairments in terms of physical functioning (p<0.01), social functioning (p<0.001), emotional role limitations (p<0.05), and general health (p<0.001) using the 36-Item Short Form Survey (SF-36).[64] HRQoL impairment is particularly pronounced if disease activity is not well-controlled.[65,66,68] Significantly poorer HRQoL has also been observed for patients with active LN compared to those with inactive disease (using LupusPRO);[66] and HRQoL worsens significantly as renal flares become more frequent (lupus impact tracker).[65] Impaired HRQoL is generally correlated with onset of symptoms (fatigue being the most burdensome) and deteriorates as patients progress to severe/advanced LN (i.e. greater renal insufficiency).[16,64,67] Treatments capable of achieving rapid renal remission may therefore be able to prevent or delay HRQoL decrements associated with disease progression.[16,64,67] However, while treatments may achieve renal remission, the systemic nature of underlying SLE means that it is particularly challenging to improve patient HRQoL due to other adversely affected organs/regions of the body.[69,70]

For these reasons, LN negatively impacts patient day-to-day activities (personal or work/study-related) and may even progress to short- or long-term disability (especially if patients develop ESRD).[54,71] Wider societal consequences are therefore expected for patients, their family, and caregivers as LN onset typically occurs at peak education/working age (18– 59).[12,31,32] In a UK-based survey of patients with SLE (n=121) and their carers (n=31; LN data not available); 52% of patients had ceased work completely, while a substantial proportion of Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

carers reported time off work (52%), negative financial implications (55%), and interference with social activities (87%).[67] In addition to the impact of LN on general day-to-day life, active LN is associated with poor maternal and foetal outcomes (Section B.1.3.5.1)[56,72] and pregnant patients have reported emotional turmoil in the form of anxiety, depression, feelings of bitterness, and worry related to renal flares during pregnancy.[73]

==> picture [44 x 10] intentionally omitted <==

Economic burden

Patients with LN are associated with substantial healthcare resource use (HRU) due to both treatment costs, and high rates of physician, inpatient and outpatient visits.[17] Patients with LN may require clinical visits for a variety of reasons that include; delivery of intravenous (IV) treatments, follow-up visits due to renal flares including repeat biopsy to assess disease progression, dialysis procedures, kidney transplantation, complications associated with longterm exposure to steroids, and/or adverse reactions to therapy.[12,27] As such, LN is known to have an even greater economic burden than patients with SLE only, particularly in terms of direct costs.[17,46,74]

Although there are no available data describing the economic burden of LN specifically in the UK, the economic burden of LN has been assessed globally across 32 clinics in 11 countries within North America, Latin America, Europe and Asia between 1999–2013 (n=1,545).[17] Tenyear cumulative direct costs were over 15-fold higher in patients with LN and poor kidney function (eGFR <30ml/min) compared to patients without LN (eGFR >60ml/min) (2015 Canadian dollars: $310,579 [approx. £207,640 in 2022] vs $19,987 [approx. £13,362 in 2022], respectively).[17] Unadjusted annual costs per patient with LN increased substantially with disease progression, increasing from $3,799 (approx. £2,542 in 2022) for patients with CKD1/2 to $50,614 (approx. £33,867 in 2022) for patients with CKD5 disease (i.e. ESRD).[17]

B.1.3.6 Clinical pathway of care

==> picture [43 x 10] intentionally omitted <==

Clinical guidelines for LN in the UK

Currently, there are no available National Institute for Health and Care Excellence (NICE) guidelines for the diagnosis and management of LN, nor have any NICE technology appraisals been completed for this indication.

In the UK, the BSR have published the only national guideline for the management of mild, moderate, and severe SLE in adults as part of a NICE-accredited process (2018). Within this guideline, the BSR recommended that patients with LN should be managed according to clinical guidelines published by the EULAR/ERA-EDTA.[13,15,27] Beyond BSR/EULAR/ERAEDTA recommendations, the most up-to-date guidelines for the treatment of LN were published by the Kidney Disease: Improving Global Outcomes (KDIGO) Glomerular Diseases Work Group (October 2021).[43] KDIGO 2021 guidelines have an international perspective, but largely reflect the BSR/EULAR/ERA-EDTA guidelines. Together, BSR/EULAR ERA-EDTA and KDIGO guidelines inform the management of LN in England and Wales.

==> picture [44 x 9] intentionally omitted <==

Diagnostic pathway

Early diagnosis and management of LN is critical to preserve kidney function and associated with improved prognosis.[28] According to BSR/EULAR/ERA-EDTA and KDIGO guidelines, kidney involvement can be identified by urinalysis or blood tests and includes patients with: glomerular haematuria, cellular casts in the urine, proteinuria >0.5 g/24 hours, spot urine Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

protein to creatinine ratio (UPCR) >500mg/g, and/or an unexplained decrease in glomerular filtration rate (GFR).[15,27,39,43] The NHS and UK Kidney Association indicate that an eGFR of <60 ml/min/1.73 m[2] may be classified as CKD; while normal eGFR is ≥90 ml/min/1.73 m[2] , or 60–89 ml/min/1.73 m[2] in the absence of kidney damage.[41,42] Patients with SLE that have persistent proteinuria ≥0.5g/day (or spot UPCR ≥500mg/g) and/or unexplained decreases in GFR should be referred for kidney biopsy to confirm and identify the extent of renal damage according to the ISN/RPS 2003 classification system (Section B.1.3.4).[27,43]

For patients with LN, glomerulonephritis due to immune complex deposits is the most common histology type, although other etiopathogenetic mechanisms may include podocytopathy or thrombotic microangiopathy.[12,39] To better differentiate among these mechanisms, patients with suspected kidney involvement may be tested for complement levels (C3 and C4); antidouble-stranded deoxyribonucleic acid (anti-dsDNA) and anti-C1q autoantibodies; and antiphospholipid antibodies, which may have prognostic implications.[27]

==> picture [44 x 9] intentionally omitted <==

Treatment pathway

In the absence of a cure, LN treatment goals include the preservation or improvement of kidney function (i.e. the normalisation/stabilisation of eGFR) and the prevention of disease progression to more advanced stages of CKD.[12,27,43] Although there are no universally accepted criteria for the level of improvement required, treatments traditionally aim to achieve a ‘CR’ in terms of a reduction of proteinuria (EULAR/ERA-EDTA definition: UPCR below 500– 700 mg/g by 12 months; KDIGO definition: UPCR 500 mg/g and/or stabilisation/improvement in kidney function [±10-15% of baseline] within 6–12 months).[27,43] In some cases, patients may require an additional 6-12 months of treatment to achieve CR (e.g. patients with nephroticrange proteinuria [UPCR ≥3000 mg/g]).[27,43]

For all patients with LN, hydroxychloroquine is recommended unless contraindicated.[28,39,43] However, patients with class III, IV or V LN may benefit from additional immunosuppressive therapy, which traditionally involves an induction phase (initial treatment) to treat patients with active LN and a follow-up maintenance phase (subsequent treatment) once the disease is adequately controlled.[13,27,43]

EULAR/ERA-EDTA and KDIGO recommend initial treatments in terms of preferred first-line treatment options and alternative treatment options which may be considered under specific circumstances (summarised in Table B.1-5).[13,27] Available treatments include immunosuppressant agents such as mycophenolate mofetil (MMF), mycophenolic acid (MPA) and azathioprine; cyclophosphamide (an immunosuppressive form of chemotherapy), and CNIs (tacrolimus or ciclosporin).[13,27,43] In addition, treatments are typically used in combination with corticosteroids during the initial treatment phase, which are then tapered to the lowest possible dose or may even be discontinued during the subsequent maintenance phase.[13,27,43]

For the initial treatment of active class III–IV LN specifically, KDIGO recommend CNIs as a triple-combination therapy with reduced-dose MMF/MPA and corticosteroids for patients who are not suitable for standard-dose MMF/MPA or cyclophosphamide (KDIGO). EULAR/ERAEDTA suggest this regimen may be particularly useful in those with nephrotic-range proteinuria (EULAR/ERA-EDTA).[13,27,43] CNIs may also be used as an initial treatment for class V LN, either with MMF/MPA and corticosteroids or with corticosteroids alone.[13,27,43] Among the two currently available CNIs (tacrolimus and ciclosporin), EULAR/ERA–EDTA guidelines and feedback from clinical experts suggest that tacrolimus is the most widely used CNI;[13,27] with

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 20 of 165

EULAR guidelines excluding ciclosporin from the recommended treatment algorithm for patients with class III–IV LN.[13]

Notably, although voclosporin has not yet received marketing authorisation in the European Union (EU), it is already licensed in the United States (US) for the treatment of active LN and KDIGO guidelines suggest that voclosporin can be added to MMF/MPA and corticosteroids as an initial therapy for up to one year.[75]

Table B.1-5. Summary of LN treatments by treatment phase and disease class according to EULAR/ERA-EDTA and KDIGO guidelines

*KDIGO-recommendation only; †EULAR/ERA-EDTA-recommendation only (KDIGO guidelines for pure class V LN are less explicit than EULAR/ERA-EDTA recommendations and generally recommend management with combined immunosuppressive treatment (i.e MMF/MPA with CYC/CNI/AZA or rituximab); ‡for corticosteroid minimisation only; §preferred if pregnancy contemplated, or following first-line CYC (EULAR/ERA-EDTA) Abbreviations: AZA = azathioprine; CNI = calcineurin inhibitor; CYC = cyclophosphamide; g = grams; GFR = glomerular filtration rate; KDIGO = Kidney Disease: Improving Global Outcomes; IV = intravenous; LN = lupus nephritis; MMF = mycophenolate mofetil; MPA = mycophenolic acid Source: Fanouriakis et al., 2020; Fanouriakis et al., 2021; KDIGO 2021[13,27,43]

Non-responding/refractory patients may be considered for treatment with MMF/MPA, cyclophosphamide, CNIs (especially tacrolimus), belimumab, and/or rituximab either as monotherapy or part ‘multitarget’ therapy.[27,43]

Despite the above recommendations, most immunosuppressive treatments are not indicated for SLE, and only two treatments are indicated for LN specifically: cyclophosphamide as a treatment of life-threatening, severe progressive forms of LN only;[63] and belimumab in combination with background immunosuppressive therapies, for the treatment of adult patients with active LN (licensed by the European Medicines Agency [EMA], but does not currently have marketing authorisation in the UK for LN).[76,77] Therefore, almost all treatments currently used for LN are prescribed off-label. A summary of treatments currently used for LN and within the scope of this appraisal is presented in (Table B.1-6).

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.1-6. Summary of therapeutic indications for current LN treatments

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 22 of 165

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 23 of 165

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 24 of 165

B.1.3.7 Unmet need

Uncontrolled, active LN causes the irreversible loss of kidney nephrons,[82] resulting in the earlier onset of ESRD and a reduction in the overall lifespan of the kidneys.[83] For this reason, patients with LN who do not respond to treatment within 12 months are around five times more likely to develop CKD (HR 5.2 [95% CI: 2.8–7.6]).[84] It is therefore critical to achieve treatment response to prevent further organ damage accrual, and improve renal prognosis.[27,28,39,84-86]

Despite currently available treatments, many patients with LN continue to develop ESRD, where almost two-thirds of patients die within five years (Section B.1.3.5.1.1).[48,50,52] Standard of care (SoC) treatment with traditional immunosuppressants (MMF or cyclophosphamide; Section B.1.3.5.1.1) is associated with sub-optimal and slow clinical response, thereby extending the length of time that a patient is exposed to active LN and nephron damage (Rovin et al., 2021).[13,27,43] Phase 3 randomised trials report CR rates of only 8.6%–30.6% for MMF and 8.1% for cyclophosphamide (Appendix M).[2,87,88] Furthermore, various studies have reported high rates of renal flare after long-term follow-up following both MMF (19% after 48 months; ~45% after 110 months [Phase 3]),[89,90] and cyclophosphamide (29% after 41 months [Phase unspecified]).[91]

CNIs have demonstrated greater efficacy than traditional immunosuppressive agents, both as monotherapy and in combination with MMF.[92-94] In particular, meta-analyses of randomised trials demonstrate that tacrolimus monotherapy is significantly more effective than cyclophosphamide at achieving CR (51% vs 31%, respectively [p=0.004]; n=225);[95] and overall response (odds ratio [OR] 2.4 [95% CI: 1.0–5.5]; n=972).[96] Similarly, a randomised trial (n=368) report significantly higher complete clearance for tacrolimus with MMF vs cyclophosphamide only (45.9% vs 25.6%, respectively; p<0.001), and significantly shorter median time to overall response (4.1 fewer weeks [95% CI: -7.9 to -2.1 weeks]).[92] Despite promising efficacy, currently available CNIs have additional safety limitations. Tacrolimus and ciclosporin are associated with key adverse events such as hypertension or kidney dysfunction; and several metabolic disorders which include glucose intolerance, dyslipidaemia, and diabetes.[93,97-99] Tacrolimus has even been shown to have a direct deleterious effect on pancreatic islets.[100] Current CNIs are further limited by their narrow therapeutic windows (i.e. the level of drug exposure required for efficacy is close to that of toxicity), so regular drug monitoring is needed in the form of blood tests during clinician visits.[93,101,102]

Corticosteroids and cyclophosphamide are also associated with toxicity and adverse outcomes, of which organ damage is a particular concern.[12,103,104] Within the UK, a 21-year prospective study of patients with SLE (n=382) found each treatment to be associated with the development of organ damage (HR 3.4 [95% CI: 2.0–5.7] and HR 2.5 [95% CI: 1.5–4.0], respectively).[105] Corticosteroids and cyclophosphamide also have a unique profile of potentially serious adverse effects. Specifically, corticosteroids may cause an increase of cardiovascular risk factors, osteoporosis, cataracts, and serious infection.[103,106] On the other hand, cyclophosphamide is associated with significantly higher frequencies of infections, leukopenia, hair loss, death, and hospital admission compared to MMF.[107] Safety concerns mean that corticosteroid-use is generally minimised where possible, and MMF may be preferred to cyclophosphamide as a first-choice treatment.[12,13,27,43]

Beyond MMF, cyclophosphamide and CNIs; rituximab is another off-label alternative that can be used as monotherapy or as an add-on therapy (with MMF or cyclophosphamide for patients

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

with active LN).[43] However, EULAR/ERA-EDTA do not currently recommend it’s use in the initial treatment setting, and rituximab failed to demonstrate a superior overall response rate (primary endpoint) to placebo in a Phase 3 trial of patients with active LN treated concomitantly with MMF/corticosteroids.[88]

In conclusion, the SoC for LN has remained largely unchanged over the past 10 years in the form of off-label immunosuppressant therapy with MMF, cyclophosphamide and corticosteroids.[27,108] Despite these treatments, patients remain at high risk of progressing to ESRD due to sub-optimal/slow clinical response, and safety concerns which limit the use of corticosteroids and cyclophosphamide.[27,108] Off-label treatment with CNIs has since shown promising efficacy; however, they are linked with additional safety limitations and are dependent on regular drug monitoring which places undue burden on patients and healthcare professionals.[1,93,97-99,102,109] Therefore, there is a critical need for novel treatments which effectively control disease activity, are more tolerable, and have the potential to minimise cyclophosphamide and/or high-dose corticosteroid use. A novel CNI that has a consistent pharmacokinetic and pharmacodynamic profile which eliminates the need for regular therapeutic drug monitoring, while retaining the efficacy benefits associated with CNI treatments, relative to traditional immunosuppressive therapy.[2]

B.1.3.8 Place of voclosporin in the treatment pathway

Voclosporin is anticipated to be used in accordance with its proposed marketing authorisation; in combination with background immunosuppressive therapies for the treatment of adult patients with active class III, IV or V (including mixed class III/V and IV/V) LN.

All active LN patients (class III, IV or V including mixed class III/V and IV/V) should be considered for treatment with voclosporin, including patients at initial diagnosis of LN, those with newly flaring disease (previously in remission), and those previously diagnosed but inadequately treated for LN.

Voclosporin should be used as a first-line alternative to MMF/MPA and cyclophosphamidebased treatments as an initial treatment of active class III, IV or V (including mixed class III/V and IV/V) LN. As the first CNI to be indicated for the treatment of LN, voclosporin should also be used ahead of other CNI-based treatments due to its improved immunosuppressive potency, tolerable safety profile, and broader therapeutic index which eliminates the need for regular therapeutic drug monitoring.[8,9,11]

B.1.4 Equality considerations

There are no known equality issues relating to the use of voclosporin in patients with LN.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 26 of 165

B.2. Clinical effectiveness

B.2.1 Identification and selection of relevant studies

A systematic literature review (SLR) was conducted to identify and summarise the available randomised controlled trial (RCT) evidence for the current and future treatment options for patients with LN. Full details of the SLR are included in Appendix D.

B.2.2 List of relevant clinical effectiveness evidence

The efficacy and safety of voclosporin has been evaluated in the pivotal Phase 3 study (AURORA 1: AUR-VCS-2016-01 [NCT03021499]),[2] as well as a follow-on Phase 3 long-term continuation study (AURORA 2: AUR-VCS-2016-02 [NCT03597464]). In addition, data is provided from a Phase 2b study with post-study long-term outcomes (AURA-LV; AUR-VCS2012-01 [NCT02141672]),[8] and a pooled analysis of Phase 3 AURORA 1 and Phase 2b AURA-LV.[110] An overview of the clinical effectiveness evidence is presented in Table B.2-1.

Table B.2-1. Clinical effectiveness evidence

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

*Defined by >10 RBCs per high powered field with dysmorphic RBC and/or RBC casts on urinalysis of a urine sample which has a minimum volume of 50 ml

Abbreviations: ACR = American College of Rheumatology; BID = twice daily; C3 = complement 3; C4 = complement 4; dsDNA = double-stranded deoxyribonucleic acid; HRQoL = health-related quality of life; LN = lupus nephritis; mg = milligrams; ml = millilitre; RBC = red blood cell; SLE = systemic lupus erythematosus

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

The efficacy and safety of voclosporin have been evaluated in a comprehensive clinical trial programme. The results of the AURORA 1 Phase 3 and AURORA 2 Phase 3 long-term continuation trials constitute the primary source of clinical evidence for this submission, along with supporting data from the AURA-LV Phase 2 study and an integrated pooled analysis of AURORA 1 and AURA-LV. A summary of methodology for AURORA 1 (Section B.2.3.1), AURORA 2 (Section B.2.4.2) and AURA-LV (Section B.2.3.3) is provided, along with supporting efficacy (Section B.2.6) and safety (Section B.2.10) data for each trial.

Across each trial, the terms “complete remission”, “complete renal remission”, “renal response” and “CRR” have been used interchangeably but share the same definition. Similarly, “partial remission”, “partial response” and “partial renal response (PRR)” have also been used interchangeably but share the same definition. For the purposes of this submission, the outcomes are henceforth referred to as “CRR” and “PRR” for consistency across all three trials.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.2.3.1 AURORA 1 Phase 3 study

==> picture [43 x 10] intentionally omitted <==

Study design and objectives: AURORA 1

AURORA 1 is a Phase 3, multicentre, double-blind, placebo-controlled, randomised trial that compared the efficacy and safety of voclosporin versus placebo, each in combination with MMF and low-dose oral corticosteroids for the treatment of patients with active LN.[2] In each treatment arm, over a period of 52 weeks, the primary objective was to assess efficacy in achieving CRR, while the secondary objective was to assess safety and tolerability of therapy in patients with active LN.[109]

An overview of AURORA 1 trial design is presented in Figure B.2-1, accompanied by a summary of methodology in Table B.2-2.

Figure B.2-1. Trial design: AURORA 1 (AUR-VCS-2016-01; NCT03021499)

==> picture [441 x 164] intentionally omitted <==

*Oral corticosteroids were tapered per protocol

Abbreviations: BID = twice-daily; MMF = mycophenolate mofetil Source: Rovin et al., 2021[2]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 29 of 165

Table B.2-2. Summary of methodology for AURORA 1 (AUR-VCS-2016-01; NCT03021499)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

• Region

• MMF use at screening and maximum MMF dose

*IV methylprednisolone (0.5 g/day for patients ≥ 45 kg, or 0.25 g/day for patients < 45 kg) once daily on days 1 and 2; followed by the commencement of oral prednisone (25 mg/day for patients ≥ 45 kg, or 20 mg/day for patients < 45 kg) on day 3. Oral prednisone was then rapidly tapered to a dose of 2.5 mg/day at Week 16, according to a protocol-specified tapering schedule. Any subsequent dose adjustments were made per investigator discretion; †According to kidney biopsy within 2 years of screening; ‡Doubling or greater increase in UPCR in the 6 months before screening was required in patients who had a kidney biopsy > 6 months before screening; §CRR is defined as a composite of UPCR of ≤0.5 mg/mg, eGFR of ≥60 ml/min/1.73[2] or no confirmed eGFR decrease of >20% from baseline, no rescue medication, and no more than 10 mg prednisone equivalent per day for ≥3 consecutive days or for ≥7 days in total during weeks 44–52

Abbreviations: ACR = American College of Rheumatology; AE = adverse event; CRR = complete renal response; ECG = electrocardiogram; eGFR = estimated glomerular filtration rate; g = grams; HRQoL = healthrelated quality of life; IV = intravenous; LN = lupus nephritis; m [2] = metres squared; mg = milligrams; MMF = mycophenolate mofetil; PROs = patient reported outcomes; PRR = partial renal response; SAE = serious adverse event; SF-36 = 36-Item Short Form Survey; UPCR = urine protein creatinine ratio

==> picture [44 x 10] intentionally omitted <==

Eligibility criteria: AURORA 1

AURORA 1 included patients diagnosed with class III, IV, or V (alone or combination with III or IV) active LN. A summary of inclusion and exclusion criteria is presented in Table B.2-3.

Table B.2-3. Inclusion and exclusion criteria: AURORA 1

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 31 of 165

*Patients with cervical dysplasia that was cervical intraepithelial neoplasia 1 but had been treated with conization or loop electrosurgical excision procedure and had a normal repeat Papanicolaou test were

Allowed; †Severe cardiovascular disease, liver dysfunction, COPD or asthma requiring steroids, bone marrow insufficiency unrelated to SLE, active bleeding disorders, or infection requiring antibiotics Abbreviations: ACR = American College of Rheumatology; BCC = basal cell carcinoma; CKD-EPI = Chronic Kidney Disease Epidemiology Collaboration; CsA = ciclosporin; eGFR = estimated glomerular filtration rate; ; LN = lupus nephritis; m [2] = metres squared; mg = milligrams; MMF = mycophenolate mofetil; SCC = squamous cell carcinoma; SLE = systemic lupus erythematosus; UPCR = urine protein creatinine ratio

==> picture [44 x 10] intentionally omitted <==

Study treatments: AURORA 1

B.2.3.1.3.1 Allocation to treatment

Patients were randomised 1:1 to the voclosporin arm or the placebo arm.[2] Randomisation was stratified by biopsy class (pure class V only vs others), previous MMF use at the time of screening (yes vs no), and region (North America vs Latin America vs Europe and South Africa vs Asia-Pacific).[2]

B.2.3.1.3.2 Treatments administered

Patients received either oral 23.7 mg voclosporin (administered as three 7.9 mg capsules) BID or matching placebo for 52 weeks.[2] In addition to the study drug (voclosporin or placebo), all patients received the following:

IV methylprednisolone (0.5 g/day for patients ≥45 kg, or 0.25 g/day for patients < 45 kg) once daily on days 1 and 2. On day 3, oral prednisone was commenced at 25 mg/day for patients ≥45 kg bodyweight, or 20 mg/day for patients <45 kg. Oral prednisone was then rapidly tapered to a dose of 2.5 mg/day at Week 16. Any subsequent dose adjustments were made per investigator discretion.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Patients who were not taking MMF prior to randomisation received 1 g/day for the first week, increasing to 2 g/day starting from day 8.

B.2.3.1.3.3 Dose modification and treatment discontinuation

Dose modification of study treatment was permitted in the case of decreased renal function, increased blood pressure, or a treatment-emergent abnormal heart rhythm (i.e. increase in QT interval duration corrected for heart rate using method of Fridericia [QTcF]).

• Decreased renal function: study treatment was withheld for any patient experiencing >30% decrease in eGFR from baseline to <60 ml/min/1.73m[2 ] until a confirmation test could be performed (at an unscheduled visit if needed). If eGFR decrease was not confirmed, study treatment was restarted at 2 capsules BID (15.8 mg voclosporin BID) and increased as tolerated with discussion with the Medical Monitor. However, if the decrease was confirmed and not due to other contributing factors, study treatment was stopped and eGFR retested for recovery within 48 hours. eGFR recovery was defined as eGFR >80% of baseline, and patients that recovered were restarted on one capsule BID [7.9mg BID] until reassessment of eGFR within 2 weeks. For patients with a <20%–≤30% reduction in eGFR to <60 mL/min/1.73 m[2 ] that was not related to other contributing factors, a confirmation test was performed within ~2 weeks (either planned study visit or an unscheduled visit) and patients were managed appropriately in consultation with the Medical Monitor by either a dose reduction (to 1 or 2 capsules BID [7.9–15.8 mg BID]) or a temporary interruption.

• Increased blood pressure: study treatment was withheld if systolic or diastolic blood pressure was >165 mmHg or >105 mmHg, respectively, and associated with symptoms of hypertension. Patients were subsequently treated per investigator local practices and best judgement, and the patient continued with all planned study visits. Study treatment could only be reintroduced following discussion with the Medical Monitor.

• Treatment-emergent increase in QTcF: any patient with a QTcF value >500 milliseconds (msec), or an increase >60msec from baseline was required to return for an unscheduled visit within 24 hours for confirmation by repeat electrocardiogram (ECG) in triplicate. If confirmed, study treatment was to be permanently discontinued and the patient was followed until the QTcF value returned to baseline (or as appropriate) or until further evaluation was not clinically indicated. If study treatment was discontinued, the patient was to remain in the study and attend all remaining planned study visits.

B.2.3.1.3.4 Concomitant therapies

A summary of permitted and prohibited concomitant therapies is shown in Table B.2-4.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-4. Concomitant therapy

*If used, patients must be on a stable dose of ACE inhibitors or ARBs for 4 weeks prior to enrolment and dose must remain stable throughout the study; †Unless approved or discussed with the Medical Monitor; ‡Concomitant use of other CYP3A45 inhibitors and inducers to be discussed with Medical Monitor Abbreviations: ACE = angiotensin-converting enzyme; AE = adverse event; ARB = angiotensin II receptor blocker; CNI = calcineurin inhibitors; CYC = cyclophosphamide; g= gram; G-CSF = granulocyte colonystimulating factor; Ig = immunoglobulin; IV = intravenous; MMF = mycophenolate mofetil; NSAIDs = nonsteroidal anti-inflammatory drug; PTS = prior to screening; PTR = prior to randomisation

==> picture [44 x 10] intentionally omitted <==

Assessments and outcomes: AURORA 1

B.2.3.1.4.1 Efficacy outcomes

The primary efficacy outcome was CRR at Week 52 assessed by the Clinical Endpoints Committee and defined as a composite of:

• UPCR of ≤ 0.5 mg/mg or less

• eGFR of ≥ 60 mL/min/1.73m[2] or no confirmed eGFR decrease of > 20% from baseline

• No administration of rescue medication

• No more than 10 mg prednisone equivalent per day for 3 or more consecutive days or for 7 or more days in total during weeks 44 through 52, just prior to the primary endpoint assessment

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Patients were disqualified from CRR if they failed both eGFR measures (i.e., confirmed eGFR <60 mL/min/1.73 m[2 ] and confirmed >20% drop from baseline) and had an associated treatment-related or disease-related adverse event (AE) that impacted eGFR. Patients who withdrew from the study prior to the Week 52 assessment and provided insufficient Week 52 data to determine response were defined as non-responders. Patients who discontinued study drug but continued to attend study visits had their data assessed for response.

The following secondary efficacy outcomes were also measured, listed below in terms of key secondary outcomes and other relevant secondary endpoints.

• Key secondary hierarchical endpoints for efficacy were analysed in order using the Hochberg step-up sequential testing procedure were (see section B.2.4.1 for more detail):[2] o Time to UPCR of ≤0.5 mg/mg

  • PRR (defined as ≥50% reduction from baseline UPCR) at Weeks 24 and 52

  • Time to 50% reduction in UPCR from baseline

  • CRR at Week 24 (based on primary endpoint definition with corticosteroid dosing assessed from Weeks 16 to 24)

• Other relevant secondary outcomes:[2]

  • Duration of UPCR ≤0.5 mg /mg

  • Proportion of patients with >30% decrease in eGFR by time point

  • Change from baseline in UPCR by time point

  • Change from baseline in serum creatinine, urine protein, and eGFR by time point

  • RR with low-dose steroids at Weeks 24 and 52

  • Change from baseline in immunology parameters (complement 3 (C3), C4, and anti-ds DNA) at weeks 24 and 52

  • Change from baseline in the safety of oestrogens in Safety of Estrogens in Lupus

  • Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index (SELENA-SLEDAI) score at Weeks 24 and 52

  • Change from baseline in HRQoL at Weeks 12, 24, and 52

  • Healthcare resource utilisation at Week 24 and 52

B.2.3.1.4.2 Efficacy assessment

Blood and urine samples contributing to efficacy and safety assessments were analysed at central laboratories using standard validated methods. Analyses included hematology, blood chemistry, coagulation, lupus markers (immunology parameters), urinalysis and eGFR.

In terms of the study primary outcome, efficacy was assessed according to the ability of study treatment to reduce the level of proteinuria (as measured by UPCR) and improve renal function (as measured by eGFR).[109,111] Urine and blood samples were collected at all study visits except Day 2 (Visit 3 of 16), for measurement of UPCR and eGFR, respectively.

The UPCR was calculated by urinalysis primarily from the first morning void (FMV), although UPCR could also be calculated from 24-hour urinalysis for the purposes of efficacy endpoints (performed at baseline, Week 24, and Week 52) should FMV samples be unavailable or invalid.

eGFR was calculated from serum creatinine results using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Patients had a minimum of two pre-study treatment eGFR measurements taken prior to dosing at baseline, and the lowest of the predose measurements was used as a marker of baseline renal function.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Disease activity was assessed using the SELENA-SLEDAI scoring system, which requires a series of tests including includes urinalysis, blood analysis, and physical examination to assess disease activity according to 24 different disease descriptors. Patients were assessed using SELENA-SLEDAI at baseline, Week 24 and Week 52.

B.2.3.1.4.3 Patient-reported outcomes

Patient-reported outcomes (PROs) were measured as secondary endpoints and included SF36 and the LupusPRO (v1.7) scores, specifically exploring the change in score from baseline at Weeks 2, 4, 8, 12, 16, 24 and 52.

B.2.3.1.4.4 Patient-reported outcome assessment

PROs were collected using the SF-36 and LupusPro (v1.7) questionnaires. SF-36 is a 36question survey designed to measure general HRQoL, while LupusPro (1.7) is a 43-question HRQoL survey specific for lupus. Patients completed surveys at the baseline visit, then at subsequent pre-planned visits at Weeks 2,4,8, 12, 24, and 52.

B.2.3.1.4.5 Safety outcomes

Safety outcomes included the collection of AE data, laboratory parameters (clinical chemistry, haematology, urinalysis), vital signs, and ECGs from baseline to week 56 (safety follow-up).

B.2.3.1.4.6 Safety assessment

Safety assessments included AEs (throughout the study), vital signs (at screening, baseline, Day 2, and Weeks 2,4,8,12,16), laboratory parameters (at screening, baseline, and Weeks 2,4,8,12,16,20,24,30,48,52, 56), 12-lead ECGs (at screening and Weeks 2,4,8,12,16,24 and 52) and physical examinations (at screening, baseline, and Weeks 24 and 52).

AEs and serious adverse events (SAEs) were defined according to International Council for Harmonisation (ICH) definitions. AEs were reported from time of patient study consent to the Safety Follow-up Visit at Week 56, while SAEs were reported from patient study consent until the final study visit, or 30 days after last study treatment administration in patients who withdrew from the study. A treatment-emergent adverse event (TEAE) was defined as any AE with onset on or after the first dose of voclosporin or placebo up to and including 30 days after the last dose of voclosporin or placebo. Patients who discontinued treatment were encouraged to attend all planned study visits up to the final Safety Follow-up Visit at Week 56, and any AEs that occurred >30 days after the last dose of voclosporin or placebo up to the Safety Follow-up were defined as post-treatment AEs.

AE intensity was assessed by the Investigator, and adjudged to be mild if the AE was easily tolerated and did not interfere with usual activity, moderate if the AE interfered with daily activity but the patient was still able to function, or severe if the AE was incapacitating and the patient was unable to work or complete usual activity.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Study population: AURORA 1

B.2.3.1.5.1 Patient disposition

Among the intention-to-treat (ITT) population (voclosporin, n=179; placebo, n=178), one patient randomised to voclosporin did not start treatment due to an AE.[2] Overall, 86.6% of all randomised patients completed the study up to Week 52 (n=162 in voclosporin arm, n=147 in placebo arm). Among patients who received at least one dose of study drug, 15 (8.4%) of 178 patients in the voclosporin group and 31 (17.4%) of 178 patients in the placebo group withdrew from the study. The most common reason for permanent study withdrawal (n ≥ 5 patients) was withdrawal of consent (voclosporin arm: n=7; placebo arm, n=14) followed by death (voclosporin arm: n=1; placebo arm: n=5) and physician decision (voclosporin arm: n=2; placebo arm: n=3).

Patients discontinuing study treatment were encouraged to remain in the study and attend all scheduled follow-up visits, up to and including the safety follow-up assessment at Week 56. Fewer patients discontinued treatment with voclosporin (33.1%) than with placebo (24.0%). The most common reason for treatment discontinuation was intolerable AEs (13.5% vs 12.8%, respectively). Patient disposition for AURORA 1 is summarised in terms of overall study withdrawals (n=30) and drug discontinuations in Table B.2-5.

Table B.2-5. AURORA 1 patient disposition

Abbreviations: AE = adverse event Source: Otsuka 2020[109]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.2.3.1.5.2 Analysis sets

The primary and hierarchical secondary efficacy analyses were completed in the ITT population, which included 179 patients randomly assigned to voclosporin and 178 patients to placebo.[2]

The safety analysis population included all patients who received at least 1 dose of study treatment.[2] One patient in the voclosporin group did not start treatment due to an AE; thus, the safety analysis population included 178 patients in the voclosporin treatment group and 178 in the placebo group.

B.2.3.1.5.3 Demographics and baseline characteristics

Baseline characteristics were similar between treatment groups.[2] Most patients (~88%) were female and the most common kidney biopsy class was pure class IV (voclosporin, 51%; placebo, 43%). The average time since initial LN diagnosis was over 4 years. A summary of demographics and baseline characteristics is shown in Table B.2-6.

Table B.2-6. AURORA 1: Baseline Demographic and Clinical Characteristics (ITT Population)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

*Analyses for race and ethnicity were post hoc; †Other include American Indian, Alaska Native, Native Hawaiian, Pacific Islander, and other or mixed races except mixed Black race; ‡Data missing for 1 patient. Percentages might not add up to 100% because of rounding.

Abbreviations: anti-dsDNA = anti-double-stranded DNA; dL = decilitre; eGFR = estimated glomerular filtration rate; ITT = intention-to-treat; mL = millilitre; MMF = mycophenolate mofetil; SELENA-SLEDAI = Safety of Estrogens in Systemic Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index; UPCR = urine protein creatinine ratio. Source: Rovin et al., 2021 [2]

B.2.3.2 AURORA 2 Phase 3 long-term continuation study

==> picture [43 x 10] intentionally omitted <==

Study design and objectives: AURORA 2

AURORA 2 is a Phase 3, multicentre, double-blind, placebo-controlled, randomised, 24-month long-term continuation study to the AURORA 1 study. Patients who completed 52 weeks of study drug treatment in the AURORA 1 study and met eligibility criteria (Section B.2.3.2.2) were allowed to continue long-term treatment as part of the AURORA 2 study.

The primary objective of AURORA 2 was to assess the long-term safety and tolerability of voclosporin compared to placebo for up to an additional 24 months following completion of treatment in the AURORA 1 study in patients with LN. All patients will continue to receive background therapy of MMF and oral corticosteroids, if applicable, starting at the same dose as at the end of the AURORA 1 study. The secondary objective was to assess the long-term efficacy of voclosporin compared to placebo for up to an additional 24 months following completion of treatment in the AURORA 1 study in patients with LN.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Figure B.2-2. Trial design: AURORA 2

==> picture [481 x 189] intentionally omitted <==

*Patients in the voclosporin arm were randomised to receive 47.4 mg/day total; 3 capsules BID (each capsule is 7.9 mg), while patients in the placebo arm received 3 placebo capsules BID (i.e. 6 capsules total per day); †Oral corticosteroids were tapered per protocol; ‡Target doses are presented; however, patients enrolled onto AURORA 2 are initiated on the same dose of study treatment, MMF, and oral corticosteroids as received at the end of the AURORA 1 study

Abbreviations: BID = twice-daily; MMF = mycophenolate mofetil Source: Otsuka et al., 2018;[112] Rovin et al., 2021[2,112]

Table B.2-7. Summary of methodology for AURORA 2 (AUR-VCS-2016-02; NCT03597464)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Abbreviations: AE = adverse event; CRR = complete renal response; eGFR = estimated glomerular filtration rate; g = grams; HRQoL = health-related quality of life; IV = intravenous; mg = milligrams; MMF = mycophenolate mofetil; PROs = patient reported outcomes; PRR = partial renal response; SELENA-SLEDAI = Safety of Estrogens in Lupus Erythematosus National Assessment - Systemic Lupus Erythematosus Disease Activity Index; SF-36 = 36-Item Short Form Survey; UPCR = urine protein creatinine ratio

==> picture [44 x 10] intentionally omitted <==

Eligibility criteria: AURORA 2

A summary of inclusion and exclusion criteria is presented in Table B.2-8.

Table B.2-8. Inclusion and exclusion criteria: AURORA 2

*Patients who had a temporary interruption and successfully restarted study drug during the AURORA 1 study were allowed to enrol into AURORA 2 with Medical Monitor approval

==> picture [44 x 9] intentionally omitted <==

Study treatments: AURORA 2

B.2.3.2.3.1 Allocation to treatment

At completion of study treatment within the AURORA 1 study (Week 52), patients meeting the required eligibility criteria continued to receive either oral voclosporin or matching placebo, as randomised in the AURORA 1 study.

B.2.3.2.3.2 Treatments administered

Patients received a maximum of either 3 capsules of voclosporin (23.7 mg BID) or matching placebo BID, as randomised into AURORA 1. Voclosporin and placebo was administered in combination with MMF and oral corticosteroids, if applicable. On enrolment into AURORA 2, treatment was commenced at the same doses given at the time of completion of the AURORA 1 study.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.2.3.2.3.3 Dose modification and treatment discontinuation

After 12 months treatment in the AURORA 2 continuation study (i.e. 24 months treatment in total), patients taking the full dose of voclosporin (23.7 mg BID [3 capsules]) were permitted to reduce the dose to 15.8 mg BID (2 capsules) at the discretion of the Investigator and after consultation with the Medical Monitor, if it was deemed that UPCR was controlled (UPCR <1.5 mg/mg).

Dose modification of study treatment was also permitted if safety concerns arose during the AURORA 2 treatment period, such as an abnormal deviation in blood pressure, a deterioration in renal function (relative to AURORA 1 baseline; see Section B.2.3.1.3.3), and gastrointestinal issues or other disturbances associated with study treatment. These risks were managed by dose reduction and temporary interruption of study treatment.

Patients were permanently discontinued from study treatment if they required treatment with IV methylprednisolone or any rescue medication other than that permitted in the protocol or experienced an unacceptable AE. Rescue medications included cyclophosphamide, rituximab, abatacept, azathioprine, eculizumab, methotrexate, and tacrolimus. If possible, patients that discontinued study treatment were expected to continue in the study and attend all study visits and assessments up to and including the final visit (Month 36 of treatment) or the early termination visit.

B.2.3.2.3.4 Concomitant therapies

Permitted and prohibited concomitant medications aligned with those specified for AURORA 1. A summary of permitted and prohibited concomitant therapies is shown in Table B.2-9.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-9. Concomitant therapy

*if used, patients must be on a stable dose of ACE inhibitors or ARBs for 4 weeks prior to enrolment and dose must remain stable throughout the study

Abbreviations: ACE = Angiotensin-converting enzyme; ARB = angiotensin II receptor blocker; CNI = calcineurin inhibitors; CYC = cyclophosphamide; g = gram; G-CSF = granulocyte colony-stimulating factor; Ig = immunoglobulin; IV = intravenous; MMF = mycophenolate mofetil; NSAIDs = non-steroidal anti-inflammatory drug

==> picture [44 x 10] intentionally omitted <==

Assessments and outcomes: AURORA 2

B.2.3.2.4.1 Efficacy outcomes

The primary outcome of AURORA 2 was to evaluate the safety of study treatment (detailed in Section B.2.3.2.4.5).

The following secondary efficacy outcomes were also measured, listed below in terms of key secondary outcomes and other relevant secondary endpoints:

• Key secondary outcomes

  • Proportion of patients in CRR that met the following criteria:

    • UPCR of ≤0.5 mg/mg

    • eGFR ≥60 mL/min/1.73 m[2] or no confirmed decrease from baseline in eGFR of >20%

    • Received no rescue medication for LN

    • Did not receive more than 10 mg prednisone for ≥3 consecutive days or for ≥7 days in total during the 8 weeks prior to the CRR assessment

  • PRR (defined as a ≥50% reduction from baseline in UPCR)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

• Renal flare as adjudicated by the Clinical Endpoints Committee (CEC)

• Extra-renal flare as adjudicated by the CEC

• Change in SELENA-SLEDAI scores from AURORA 1 baseline

• Change in UPCR, eGFR, urine protein, and serum creatinine from AURORA 1 baseline

• Other relevant secondary outcomes:

  • Change in immunology parameters (C3, C4, and anti dsDNA) from AURORA 1 baseline

  • o Change in HRQoL (SF-36) from AURORA 1 baseline

  • Healthcare resource utilisation

As an additional exploratory endpoint, kidney biopsies were performed to evaluate renal change in immune markers, histology, and transcriptomics, from the pre-study biopsy in AURORA 1, comparing it to a repeat biopsy conducted within approximately 6 months of starting treatment in this study.

B.2.3.2.4.2 Efficacy assessment

In line with AURORA 1, blood and urine samples contributing to efficacy and safety assessments were analysed at central laboratories using standard validated methods. Analyses included haematology, blood chemistry, coagulation, lupus markers (immunology parameters), urinalysis and eGFR.

CRR (according to eGFR and UPCR levels) was assessed every 3 months (Month 12, 15, 18, 21, 24, 27, 30, 33, 36) at all visits up to and including an additional Safety Follow-up visit at month 37 following completion of study treatment. eGFR and UPCR was assessed in accordance with AURORA 1 methodology (Section B.2.3.1.4.2).

Disease activity was also measured using SELENA-SLEDAI, with assessments conducted at Month 12, 18, 24, and 36 of the study.

B.2.3.2.4.3 Patient-reported outcomes

PROs were measured as a secondary endpoint, whereby SF-36 scores were recorded as a change from AURORA 1 baseline.

B.2.3.2.4.4 Patient reported outcome assessment

PROs using SF-36 were collected at study visits occurring every 6 months at Month 12, 18, 24, 30, and 36.

B.2.3.2.4.5 Safety outcomes

The primary outcome of AURORA 2 was to assess the AE profile and routine biochemical and haematological assessments for up to 36 months of study treatment (i.e. 12 months within AURORA 1 and 24 additional months within AURORA 2).

B.2.3.2.4.6 Safety assessment

Safety assessments included AEs (throughout the study), vital signs and laboratory parameters (at Months 12, 15, 18, 21, 24, 27, 30, 33, 36, 37), 12-lead ECGs (at Months 12, 18, 24, 36) and physical examinations (at Months 12 and 36).

AEs and SAEs were reported in accordance with AURORA 1 methodology (i.e. using ICH definitions; see Section B.2.3.1.4.6), with AEs reporting occurring from time of patient study consent until the final Safety Follow-up visit at Month 37 while SAEs were reported from patient Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

study consent until 30 days following final Safety Follow-up visit or 30 days after last study treatment administration in patients who withdrew or discontinued prior to study completion.

==> picture [44 x 9] intentionally omitted <==

Study population: AURORA 2

B.2.3.2.5.1 Patient disposition

Of the 357 patients enrolled in AURORA 1, a total of 216 patients (60.5%) continued to receive blinded treatment in AURORA 2: including 116 of 179 (64.8%) patients from the voclosporin arm and 100 of 178 (56.2%) patients from the placebo arm.

Of the remaining 141 patients that enrolled in AURORA 1 but did not enroll in AURORA 2; xx patients had withdrawn from AURORA 1 prematurely (xx voclosporin-treated and xx placebotreated), xx patients (xx in each arm) who completed AURORA 1 had permanently discontinued study treatment,xx patients did not enter AURORA 2 for administrative reasons (such as health authority approval not received in time or a decision for a site/country to not participate), and x patients did not give consent due to planned pregnancy or moving home. Reasons for non-participation were not captured for the remaining xx patients who were potentially eligible but did not enter AURORA 2. The xxx patients who did not enroll in AURORA 2 were not followed beyond the previous AURORA 1 safety follow-up visit at Week xx.

Among the xxx patients that enrolled in AURORA 2, xxx completed the study (xxxx%), with slightly more patients in the voclosporin arm (xxx patients, [xxxx%]) than the placebo arm (xx patients [xxxx%]) reaching the end of the study at Month 36. All patients received at least one dose of study treatment. Therefore, a total of xx patients withdrew prematurely from the study (xx patients in the voclosporin arm [xxxx%] and xx patients in the placebo arm [xxxx%]). In both arms, the most common reason for early permanent study withdrawal was withdrawal of consent (xxxx patients [xxx%]) in the voclosporin arm and xxxx [xxx%]) patients in the placebo arm). Notably, only patients in the placebo arm permanently withdrew from the study early due to death (x patients, [xxx%]) or intolerable AEs (x patients, [xxx%]).

Patients discontinuing study treatment were encouraged to remain in the study and attend all scheduled follow-up visits, up to and including the safety follow-up assessment at Month 37. Fewer patients discontinued treatment with voclosporin (xxxx%) than with placebo (xxxx%). The most common reason for treatment discontinuation was intolerable AEs, which was recorded more often in the placebo arm than the voclosporin arm (xxxx% versus xxx%, respectively). Patient disposition for AURORA 2 is summarised in terms of overall study withdrawals (n=xx) and drug discontinuations in Table B.2-10.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 45 of 165

Table B.2-10. AURORA 2 patient disposition

==> picture [452 x 277] intentionally omitted <==

----- Start of picture text -----
Parameters Voclosporin Placebo Overall
(n=116) (n=100) (N=216)
Study withdrawals, n (%) xxxxxxxxx xxxxxxxxx xxxxxxxxx
Intolerable AE xxxxxxx xxxxxxx xxxxxxx
Death xxxxxxx xxxxxxx xxxxxxx
Lost to follow-up xxxxxxx xxxxxxx xxxxxxx
Physician decision xxxxxxx xxxxxxx xxxxxxx
Pregnancy xxxxxxx xxxxxxx xxxxxxx
Protocol non-compliance xxxxxxx xxxxxxx xxxxxxx
Withdrawal of consent xxxxxxx xxxxxxx xxxxxxx
Lack of efficacy xxxxxxx xxxxxxx xxxxxxx
Treatment discontinuation, n (%) xxxxxxxxx xxxxxxxx xxxxxxxxx
Intolerable AE xxxxxxx xxxxxxxxx xxxxxxxx
Death xxxxxxx xxxxxxx xxxxxxx
Lost to follow-up xxxxxxx xxxxxxx xxxxxxx
Physician decision xxxxxxx xxxxxxx xxxxxxx
Prohibited medication required xxxxxxx xxxxxxx xxxxxxx
Pregnancy xxxxxxx xxxxxxx xxxxxxx
Protocol non-compliance xxxxxxx xxxxxxx xxxxxxx
Withdrawal of consent xxxxxxx xxxxxxx xxxxxxx
Lack of efficacy xxxxxxx xxxxxxx xxxxxxx
Other xxxxxxx xxxxxxx xxxxxxx
Study withdrawals
Treatment discontinuations
----- End of picture text -----

Abbreviations: AE = adverse event Source: Otsuka 2022[113]

B.2.3.2.5.2 Analysis sets

Of the 216 enrolled patients to the AURORA 2 study, all patients were included in both the ITT and safety analyses.

B.2.3.2.5.3 Demographics and baseline characteristics

Baseline characteristics of the patients enrolled in AURORA 2 were well balanced across the two treatment arms and were consistent with the AURORA 1 population (see Table B.2-11). The median age was xx years and most patients (xx%) were female. The majority of patients were White (xx%) or Asian (xx%) and there were more Black patients in the voclosporin arm (xx%) than the placebo arm (x%).

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-11. AURORA 2: Baseline Demographic and Clinical Characteristics

*Asian race includes Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese and Other Asian. Abbreviations: eGFR = estimated glomerular filtration rate; mg = milligram; mL = millilitre; NR = not reported; SD = standard deviation; UPCR = urine protein creatinine ratio Source: Otsuka 2022[113]

B.2.3.3 AURA-LV Phase 2 study

==> picture [43 x 9] intentionally omitted <==

Study design and objectives: AURA-LV

AURA-LV is a Phase 2, multicentre, double-blind, placebo-controlled, randomised trial of 2 doses of voclosporin versus placebo added to MMF and rapidly tapered low-dose oral corticosteroids for the treatment of patients with active LN.[8] The primary objective of AURALV was to evaluate whether voclosporin added to background therapy was more effective in inducing CRR at 24 weeks compared to background therapy alone in patients with active LN. Secondary objectives were to evaluate the efficacy, safety, and tolerability of voclosporin compared with placebo after 48 weeks of treatment.

An overview of AURA-LV trial design is presented in Figure B.2-3, accompanied by a summary of methodology in Table B.2-12.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Figure B.2-3. AURA-LV: Trial Design (AUR-VCS-2012-01; NCT02141672)

==> picture [357 x 260] intentionally omitted <==

*Oral corticosteroids were tapered per protocol. Abbreviations: BID = twice daily; MMF = mycophenolate mofetil Source: Rovin et al., 2019[8]

Table B.2-12. Summary of methodology for AURA-LV (AUR-VCS-2012-01; NCT02141672)

*CRR defined as a composite of a decrease in UPCR to ≤0.5 mg/mg, and eGFR of ≥60 mL/min/1.73 m[2] or no eGFR decrease of ≥20% from baseline

Abbreviations: BID = twice daily; CRR = complete renal response; IV = intravenous; LN = lupus nephritis; MMF = mycophenolate mofetil; PRR = partial renal response; UPCR = urine protein creatinine ratio Source: Rovin et al., 2019; [8] Otsuka et al., 2018 [114]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Eligibility criteria: AURA-LV

A summary of inclusion and exclusion criteria of AURA-LV is presented in Table B.2-13.

Table B.2-13. Inclusion and exclusion criteria: AURA-LV

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

mofetil; MPA = mycophenolic acid; SCC = squamous cell carcinoma; SLE = systemic lupus erythematosus; UPCR = urine protein creatinine ratio

==> picture [44 x 10] intentionally omitted <==

Study treatments: AURA-LV

B.2.3.3.3.1 Allocation to treatment

Patients were randomised 2:2:1:1 to low-dose voclosporin (23.7 mg BID), high-dose voclosporin (39.5 mg BID), low-dose matched placebo, and high-dose matched placebo, respectively.[8] Randomisation was stratified by biopsy class (Class V vs others) and by MMF use at screening.

B.2.3.3.3.2 Treatments administered

Patients received either twice-daily low-dose voclosporin (23.7 mg [3 capsules] BID) or highdose voclosporin (39.5 mg [5 capsules]), or a low- or high-dose matched placebo for up to 48 weeks.[8]

In addition to the study drug (voclosporin or placebo), all patients received the following:[8]

• An initial treatment with 0.5 g IV methylprednisolone, followed by a reducing taper of oral corticosteroid (prednisone) to a target of 2.5 mg/day by Week 16.

• • Background therapy with MMF at a target dose of 2 g/day

B.2.3.3.3.3 Dose modification and treatment discontinuation

Dose-modification instructions were included in the protocol for cases of deterioration in renal function, increased blood pressure, or increase in QTcF.

• Decreased renal function: During the treatment period, any patient with a >10-20% reduction in eGFR compared to baseline had potential contributing factors ruled out and appropriate corrective action taken. Any patient with a confirmed >20-30% reduction in eGFR compared to baseline not due to potential contributing factors had the dose of study drug reduced by 1 capsule BID (15.8 mg/day). A repeat eGFR was performed within 2 weeks (at planned study visit if any or an unscheduled visit). If the eGFR remained >2030% below the baseline value, then the dose was reduced by a further 1 capsule BID (15.8 mg/day). A maximum of two dose reductions were permitted in total throughout the study. If the patient’s eGFR did not return to within 20% of the baseline value after two dose reductions, Investigators were instructed to discontinue the patient’s voclosporin treatment permanently. Any patient that experienced a >30% decrease in eGFR from baseline, not within normal range, had study drug withheld until a repeat test could be performed (unscheduled visit to be completed). If the decrease was confirmed and not due to potential contributing factors, the case was discussed with the Medical Monitor and study drug was discontinued permanently. If the decrease in eGFR >20% was confirmed and considered due to potential contributing factors, corrective action was taken and the patient reassessed within 2 weeks.

• Increased blood pressure: If on any study day, systolic blood pressure was >165 mmHg or diastolic blood pressure was >105 mmHg and was associated with symptoms of hypertension, the study drug was discontinued, and the patient treated as per Investigator local practices and best judgment.

• Treatment-emergent increase in QTcF: In the event that a patient had a QTcF value exceeding 500 msec or an increase of >60 msec from baseline, the Medical Monitor was

• Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

to be informed. The patient was required to return for an unscheduled visit to have the ECG repeated. If any of the repeat measurements of QTcF were >500 msec or increased >60 msec from baseline, the study drug was to be permanently discontinued and the patient followed until the QTcF value either returned to baseline or until, in the judgment of the Investigator, further evaluation was not clinically indicated.

B.2.3.3.3.4 Concomitant therapies

A summary of permitted and prohibited concomitant therapies is shown in Table B.2-14.

Table B.2-14. Concomitant therapy: AURA-LV

*if used, patients must be on a stable dose of ACE inhibitors or ARBs for 4 weeks prior to enrolment and dose must remain stable throughout the study; †Unless approved or discussed with the Medical Monitor; ‡concomitant use of other CYP3A45 inhibitors and inducers to be discussed with Medical Monitor Abbreviations: ACE = Angiotensin-converting enzyme; ARB = angiotensin II receptor blocker; CNI = calcineurin inhibitors; CsA = ciclosporin A; CYC = cyclophosphamide, IV = intravenous; MMF = mycophenolate mofetil; PTS = prior to screening; RTX = rituximab; TAC = tacrolimus

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Assessments and outcomes: AURA-LV

B.2.3.3.4.1 Efficacy outcomes

The primary outcome was the number of patients showing CRR at Week 24 in the full analysis set, defined as follows:

• Confirmed decrease in proteinuria as defined by a UPCR of ≤0.5 mg/mg, and;

• Normal or stable renal function defined as no confirmed eGFR <60 mL/min/1.73 m[2] or no confirmed decrease from baseline in eGFR of ≥20%.

Patients were not defined as achieving CRR if they received rescue medication for LN or >10 mg prednisone for >3 consecutive days or >7 days total from 56 days prior to response assessment until the time of the remission assessment. The use of rescue medications was adjudicated prior to unblinding of the study. Patients who withdrew early from the study were excluded from being defined as achieving CRR; however, patients who discontinued study drug but continued study visits were considered for CRR.

The following secondary efficacy outcomes were also measured, listed below in terms of key secondary outcomes and other relevant secondary endpoints:

• Key secondary outcomes

  • CRR at Week 24 and Week 48 using 24-hour urine measurements (instead of FMV)

  • CRR in the presence of low dose steroids at Week 24 (i.e. ≤5 mg prednisone for ≥8 weeks) and Week 48 (in which no UPCR confirmation was required and patients received ≤5 mg prednisone for ≥12 weeks)

  • Time to CRR

  • Time to sustained CRR (defined as the first occurrence of CRR that was sustained through Week 48)

  • Time to (and proportion achieving) sustained early CRR (defined as CRR that occurred on or before Week 24 and was sustained through Week 48)

  • Duration of CRR (in months)

  • PRR at Week 24 and Week 48

  • Time to PRR

  • Time to sustained PRR (defined as the first occurrence of PRR that was sustained through Week 48)

  • Time to (and proportion achieving) sustained early PRR (defined as partial remission that occurred on or before Week 24 and was sustained through Week 48)

• Other relevant secondary outcomes:

  • Change from baseline in UPCR at Week 24 and Week 48

  • Change from baseline in eGFR, serum albumin, urine protein, and serum creatinine at each time point

  • Proportion of patients with active urinary sediment (defined by >10 red blood cells (RBC)/high-powered field with dysmorphic RBC and/or RBC casts on urinalysis of a urine sample which had a minimum volume of 50 mL) at each visit

  • Change from baseline in the SELENA-SLEDAI score at Week 24 and Week 48

  • Change from screening in immunology parameters (C3, C4, and anti-double-stranded DNA (anti-dsDNA)) and biomarkers (cardiolipin antibodies) at Weeks 12, 24, and 48

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 52 of 165

B.2.3.3.4.2 Efficacy assessment

Blood and urine samples contributing to efficacy and safety assessments were analysed at central laboratories using standard validated methods

Confirmed decreases in proteinuria and eGFR were defined as those where the measurement met the specified criterion at two consecutive measurements at least 3 days apart, one of which being the windowed Week 24 assessment. eGFR was assessed at all screening and all other visits except baseline up to Week 50.

UPCR values were determined using FMV urinalysis samples. In the event that the Investigator determined that the screening, Week 24, or Week 48 FMV urinalysis sample was invalid, the UPCR values from the 24-hour urinalysis at the corresponding visit were used instead of the FMV urinalysis in the calculation of efficacy endpoints. FMV analysis results were still used for all other timepoints.

B.2.3.3.4.3 Safety outcomes

The safety and tolerability of two doses of voclosporin was assessed as a secondary objective over 48 weeks compared to placebo in patients with active LN. Safety outcomes included the collection of AE data, laboratory parameters (clinical chemistry, haematology, urinalysis), physical examinations, vital signs, and blood pressure management and cardiovascular safety.

B.2.3.3.4.4 Safety assessment

Safety outcomes included the collection of AE data, laboratory parameters (clinical chemistry, haematology, urinalysis), physical examinations, vital signs, and blood pressure management and cardiovascular safety.

A summary of MMF exposure, including mean daily dose and incidence of non-protocol specified dose modifications were reported. Safety analyses were based on TEAEs, defined as AEs that occurred or increased in intensity after the first dose of study drug, up to 14 days after study completion/withdrawal.

Symptomatic increased blood pressure was identified in the study protocol as a medically important event and was required to be reported as an SAE. All patients identified were assessed for the seriousness of their symptoms, their relationship to study drug treatment, and association of symptoms related to increased blood pressure. Additional AEs of special interest considered in this study included reduced renal function, QTcF prolongation, and malignancies.

A retrospective high-level safety follow-up of the LN disease status of patients was performed approximately six to nine months following either their last study visit or last dose of study drug was also conducted.

==> picture [44 x 9] intentionally omitted <==

Study population: AURA-LV

B.2.3.3.5.1 Patient disposition

Among the ITT population, 73 (82%) patients randomised to low-dose voclosporin and 80 (90.9%) patients randomised to high-dose voclosporin completed 48 weeks of treatment compared to 70 (79.5%) patients randomised to placebo.[8]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Reasons for discontinuation were generally similar across treatment groups except for discontinuation due to death.[8] A higher proportion of patients in the low-dose voclosporin group (n=10, 11.2%) died during the study compared with the high-dose voclosporin (n=2, 2.3%) or placebo groups (n=1, 1.1%). Nine of the 13 deaths occurred in the first 2 months of study enrolment, and more than half of deaths (n=7 of 13) were among patients at 2 sites in Bangladesh. Two-fold more patients were randomised to low-dose voclosporin than placebo at these 2 sites, which may possibly be relevant to the imbalance of deaths.

B.2.3.3.5.2 Analysis sets

The ITT population included 89 patients randomised to low-dose voclosporin, 88 to high-dose voclosporin, and 88 patients to placebo (44 each to low-dose and high-dose matched placebo).[8]

B.2.3.3.5.3 Demographics and baseline characteristics

Treatment groups were generally similar with respect to demographics and clinical characteristics, though there were some numeric differences between treatment groups (Table B.2-15).[8] A higher proportion of women (92.0%) were randomised to high-dose voclosporin than to low-dose voclosporin (85.4%) or placebo (83.0%). There was also a higher proportion of patients with Class III + V or IV + V LN randomised to low-dose voclosporin (23.6%) than to the high-dose voclosporin (12.5%) or placebo groups (18.2%). The placebo group included a higher proportion of White patients (47.7%) compared with the voclosporin groups (low-dose 33.7%; high-dose 40.9%). These differences did not appear related to the study outcomes.

Table B.2-15. AURA-LV: Baseline Demographic and Clinical Characteristics (ITT Population)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

*Bangladesh, Sri Lanka (Asian-Indian Subcontinent) + Hong Kong, Korea, Philippines, Singapore, Taiwan, Thailand (Asian—other); †Number evaluated was 87

Abbreviations: eGFR = estimated glomerular filtration rate; ITT = intention-to-treat; LN = lupus nephritis; MMF = mycophenolate mofetil; SD = standard deviation; UPCR = urine protein to creatinine ratio Source: Otsuka 2018;[114] Rovin et al., 2019 [8]

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

B.2.4.1 AURORA 1

==> picture [43 x 9] intentionally omitted <==

Determination of sample size

AURORA 1 sample size was estimated to ensure that a two-group continuity corrected Chi square test (two-sided p=0.05) will have 80% power to detect the difference between a CRR rate of 20.0% in the placebo arm, and 34.4% in the voclosporin arm (OR: ~2.1), indicative of a clinically relevant treatment effect.[111]

On this basis, a sample size of 324 patients (162 patients per treatment arm) was determined for enrolment.

==> picture [44 x 9] intentionally omitted <==

Study group definitions

Three analysis populations were defined for AURORA 1:[109,111]

• Intent-to-Treat (ITT) population comprised all patients who were randomised to treatment.

  • All efficacy analyses were performed on the ITT population.

• Per-Protocol (PP) population comprised all patients from the ITT population who did not have any major protocol violations (i.e., deviations considered to have a serious impact on the efficacy results). Supplementary analyses were performed on the PP population to assess the impact of protocol deviations on CRR.

• Safety population comprised all randomised patients who took at least one dose of study treatment. Patients who received treatment from more than one arm were to be assigned to the voclosporin arm. Safety analyses were performed on the safety population.

==> picture [44 x 10] intentionally omitted <==

Efficacy analyses

An overall type 1 (false-positive) error rate of 5% was maintained so that no statistical significance for the key secondary efficacy endpoints was claimed unless the primary efficacy endpoint (CRR at Week 52) was statistically significant at the 5% level (i.e. p<0.05). Key

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

secondary efficacy endpoints were tested using the Hochberg step-up procedure to adjust for multiple comparisons amongst key secondary endpoints and maintain the overall type 1 (falsepositive) error rate of 5%.[109,115]

B.2.4.1.3.1 Renal response

Logistic regression models were used to conduct the primary efficacy analysis (CRR at Week 52) and secondary efficacy analyses of CRR at Week 24, PRR at Week 24/52, and CRR with low-dose steroids at Week 24/52. Logistic regression models incorporated baseline variables within the model as appropriate.[109]

B.2.4.1.3.2 Time to event endpoints

Time to event endpoints for time to UPCR <0.5 mg/mg, 50% reduction in UPCR and duration of UPCR <0.5 mg/mg were measured from baseline as the number of weeks from day of randomisation to the day of the event. Patients who did not experience an event were censored on the day of their last assessment of UPCR. Time-to-event was estimated using Kaplan-Meier methodology and analysed by comparing the survivor function between treatment arms. A Cox’s proportional hazards model was performed to assess the significance of the differences between treatment arms, incorporating terms for treatment, baseline UPCR, biopsy class, and MMF use at baseline.[109]

B.2.4.1.3.3 Change from baseline endpoints

Change from baseline endpoints (UPCR, serum creatinine, urine protein, HRQoL [SF-36 and LupusPRO], and SELENA-SLEDAI score) were analysed using a mixed effect model repeated measures (MMRM) analysis with treatment arm, visit, treatment by visit interaction, biopsy class, MMF use at baseline, region, and baseline UPCR included as covariates in the model.[109]

A patient recording a clinically significant >30% decrease in eGFR from baseline at any two consecutive time points was considered a confirmed eGFR drop patient at the earlier time point. The proportion of patients with a confirmed eGFR drop at each visit was compared using the Chi Square test.[109]

==> picture [44 x 10] intentionally omitted <==

Safety analyses

Descriptive statistics of safety were collected using MedDRA Version 20.0, and medications were categorised using the World Health Organisation Anatomical Therapeutic Chemical (WHO ATC) system. All TEAEs (including drug-related), SAEs and drug-related SAEs were tabulated by System Organ Class and preferred term (PT).

B.2.4.2 AURORA 2

==> picture [43 x 9] intentionally omitted <==

Determination of sample size

Patients from AURORA 1 entered AURORA 2 to provide the opportunity of an additional 24 months of treatment (a total of 36 months of treatment); therefore, no sample size calculation was required.[112]

==> picture [44 x 9] intentionally omitted <==

Study group definitions

Two analysis populations were defined for AURORA 2:[109,111]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

• Intent-to-Treat (ITT) population comprised all patients who consented to AURORA 2 treatment. Patients were analysed based on the treatment they were randomised to in AURORA 1. All efficacy analyses were performed on the ITT population.

• Safety population comprised all randomised patients who took at least one dose of study treatment. Patients who received treatment from more than one arm were to be assigned to the voclosporin arm. Safety analyses were performed on the safety population.

==> picture [44 x 10] intentionally omitted <==

Efficacy analyses

B.2.4.2.3.1 Renal response and renal/extra-renal flares

Logistic regression models were used to conduct the secondary efficacy analysis (RR and PRR at Months 12, 18, 24, 30, 36). Logistic regression models incorporated baseline variables within the model as appropriate (terms for treatment group, baseline UPCR, biopsy class, and MMF use at baseline).[109] The proportion of patients with renal and extra-renal flares (as adjudicated by the CEC) were analysed in a similar manner.[109]

B.2.4.2.3.2 Change from baseline endpoints

Change from AURORA 1 baseline endpoints (UPCR, serum creatinine, urine protein); and AURORA 2 change in HRQoL (SF-36) and SELENA-SLEDAI were analysed using a MMRM analysis with treatment arm, visit, treatment by visit interaction, biopsy class, MMF use at baseline, region, and baseline UPCR included as covariates in the model.[109]

==> picture [44 x 10] intentionally omitted <==

Safety analyses

Descriptive statistics of safety were collected using MedDRA Version 20.0, and all TEAEs (including drug-related), SAEs and drug-related SAEs were tabulated by System Organ Class and PT.

B.2.4.3 AURA-LV

==> picture [43 x 10] intentionally omitted <==

Determination of sample size

AURA-LV sample size was estimated using a two-sided alpha level of 0.05. Assuming a CRR of 20% in the placebo arm and 41% in the either of the two voclosporin arms, a sample size of 86 patients per arm was (n=258) determined to provide 81% power to detect clinically relevant significant difference between treatment arms (OR = 2.78). 86 patients per arm was also deemed sufficient to estimate the rates of AEs to an acceptable level of precision (i.e. an event with an incidence of 15% would have a 95% CI of 8.3–24.5%, and an event with an incidence of 6% would have a 95% CI of 1.9 –13.0%).[114] On this basis, a sample size of 258 patients were determined for enrolment.

==> picture [44 x 10] intentionally omitted <==

Study group definitions

Four analysis populations were defined for AURA-LV:[114]

• Full analysis set (FAS) comprised all patients who were randomised to treatment, received at least one dose of study drug, and had at least one post-baseline assessment. The primary population for efficacy analyses was the FAS.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

• Per-Protocol Set (PPS) population comprised all patients from the FAS population who did not have any major protocol violations (e.g. lack of compliance to protocol-specified steroid tapering schedule, failure to meet inclusion/exclusion criteria, and use of prohibited concomitant medication). Key efficacy analyses were repeated for the PPS as supportive analyses.

• Modified Intent-to-Treat (mITT) comprised all randomised patients who took at least one dose of study treatment. Patients who received treatment from more than one arm were to be assigned to the voclosporin arm. Safety analyses were performed on the safety population. Key efficacy analyses were repeated for the mITT as supportive analyses.

• Safety Set (SS) comprised all randomised patients who took at least one dose of study treatment. Safety analyses were performed on the safety population.

==> picture [44 x 9] intentionally omitted <==

Efficacy analyses

All statistical tests were two-sided with no adjustments for multiple comparisons (level of significance p<0.05).[114]

B.2.4.3.3.1 Renal response

Logistic regression models were used to conduct the primary efficacy analysis (CRR at Week 24) and secondary efficacy analyses of CRR at Week 48. PRR at Week 24/48, and CRR with low-dose steroids at Week 24/48. Logistic regression models incorporated baseline variables within the model as appropriate.[114]

B.2.4.3.3.2 Time to event endpoints

Time to CRR (UPCR <0.5 mg/mg) and PRR was measured from baseline as the number of days from randomisation to the day of the event. Time to sustained CRR/PRR and sustained early CRR/PRR (beginning ≤Week 24 assessment window) was measured from baseline to CRR/PRR that was sustained through the Week 48 visit. Each time to event endpoint was estimated using Kaplan-Meier methodology and Cox’s proportion hazards model. A two-sided log-rank test was performed to assess the significance of differences between the two treatment groups.[114]

B.2.4.3.3.3 Change from baseline endpoints

Change from baseline endpoints (UPCR, eGFR, serum albumin, urine protein, and SELENASLEDAI score) were analysed using analysis of covariance (ANCOVA) models adjusted as appropriate.[114]

==> picture [44 x 10] intentionally omitted <==

Safety analyses

Safety data were summarised for the SS, with the exception of eGFR data which was analysed for the FAS and included as part of efficacy and safety analyses.

Descriptive statistics of safety were collected using Medical Dictionary for Regulatory Activities (MedDRA) Version 17.0, and all TEAEs (including drug-related), SAEs and drug-related SAEs were tabulated by SOC and PT.[114]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 58 of 165

B.2.5 Critical appraisal of the relevant clinical effectiveness evidence

Table B.2-16. Quality assessment of AURORA 1

Table B.2-17. Quality assessment of AURORA 2

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-18. Quality assessment of AURA-LV

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Abbreviations: CSR = clinical study report; FAS = full analysis set; GDP = gross domestic product; mITT = modified intent-to-treat; MMF = mycophenolate mofetil; SS = safety set; TEAE = treatment-emergent adverse event Sources: Otsuka 2018[114]

B.2.6 Clinical effectiveness results of the relevant trials

B.2.6.1 AURORA 1 Phase 3 study

AURORA 1 assessed the efficacy and safety of voclosporin compared with placebo in achieving renal response after 52 weeks of therapy in patients with active LN. A total of 357 eligible patients were randomised into two groups with well-balanced demographic characteristics: 178 to the placebo arm and 179 to the voclosporin arm.

The study met its primary objective, demonstrating that treatment with voclosporin results in a clinically meaningful and statistically significant higher renal response rate compared to placebo.

==> picture [43 x 10] intentionally omitted <==

Complete Renal Response at Week 52 (primary endpoint)

In AURORA 1, significantly more patients treated with voclosporin than with placebo achieved a CRR at Week 52 (73 [40.8%] vs 40 [22.5%] patients; OR 2.65; [95% CI 1.6, 4.3]; p<0.0001).[2] The absolute difference between groups for achieving a CRR was 18% in favour of voclosporin; therefore, the number-needed-to-treat (NNT) with voclosporin is 6 individuals with active LN.[2]

All composite measures of CRR occurred more frequently in patients treated with voclosporin than with placebo, but only UPCR ≤ 0.5 mg/mg was significantly different (Table B.2-19).

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-19. AURORA 1: Summary of CRR (primary endpoint) and composites of CRR

Abbreviations: AE = adverse event; CI = confidence interval; CRR = complete renal response; eGFR = estimated glomerular filtration rate; LN = lupus nephritis; OR = odds ratio; UPCR = urine protein creatinine ratio Source: Otsuka 2020;[109] Rovin et al., 2021 [2]

==> picture [44 x 10] intentionally omitted <==

Time to UPCR of ≤0.5 mg/mg (secondary endpoint)

More patients in the voclosporin arm achieved UPCR ≤0.5 mg/mg versus the placebo arm, (64.8% vs 43.8%) and the time to UPCR ≤0.5 mg/mg was also significantly shorter with voclosporin (median time: 169 days vs 372 days; HR 2.0; [95% CI: 1.5, 2.7]; p<0.001; Figure B.2-4).[2]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Figure B.2-4. AURORA 1: Probability of UPCR of ≤0.5 mg/mg

==> picture [447 x 308] intentionally omitted <==

Abbreviation: CI = confidence interval; HR = hazard ratio; UPCR = urine protein creatinine ratio Percentiles are Kaplan-Meier estimates. The HRs are from a Cox’s proportional hazards model with covariates for treatment group, baseline UPCR, biopsy class, mycophenolate mofetil use at baseline, and region. Source: Rovin et al., 2021[2]

==> picture [44 x 10] intentionally omitted <==

Complete Renal Response at Week 24 (secondary endpoint)

The numbers of patients achieving renal response at Week 24 per CEC adjudication were lower than the numbers achieving response at Week 52. Consistent with the primary endpoint, the proportion of patients achieving an adjudicated renal response at Week 24 was significantly higher in the voclosporin arm than the placebo arm (32.4% vs 19.7%; OR 2.23; [95% CI: 1.3, 3.7]; p=0.002).[2]

==> picture [44 x 10] intentionally omitted <==

Partial Renal Response at Weeks 24 and 52 (secondary endpoint)

Consistent with the results for renal response, more patients in the voclosporin arm achieved a PRR (defined as a 50% reduction from baseline in UPCR) at Week 24 and Week 52 (Table B.2-20). In both arms, PRR was achieved by Week 24 in the majority of patients who responded. The response rate of approximately 50% in the placebo arm demonstrates that the MMF and steroid regimen used in the study is effective in reducing UPCR; however, a greater number of patients responded in the voclosporin arm.

Table B.2-20. PRR at Weeks 24 and 52

Abbreviations: CI = confidence interval; OR = odds ratio; PRR = partial renal response

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Source: Rovin et al., 2021[2]

==> picture [44 x 10] intentionally omitted <==

Time to 50% Reduction in UPCR (secondary endpoint)

A 50% reduction in UPCR from baseline at any time during the study was achieved by 96.6% of patients treated with voclosporin compared with 75.8% of patients receiving placebo. The time taken to reach a 50% reduction in UPCR was significantly shorter for the voclosporin arm than the placebo arm (HR 2.05; 95% CI: 1.6, 2.6; p<0.001). Median time to 50% reduction in UPCR was 29 days for voclosporin versus 63 days for placebo. Similar results were seen when using the lowest available pre-dose UPCR measurement as baseline.[2]

Consistent with the time to UPCR ≤0.5 mg/mg, the difference between the two treatment arms in the time to 50% reduction in UPCR was apparent within the first month of treatment and was sustained throughout the study (Figure B.2-5). The Kaplan-Meier curve shows that a small number of patients in the placebo arm achieved a 50% reduction in UPCR late on the study (beyond Day 350). However, most patients in the voclosporin arm achieved this response earlier; 6 and 38 patients were classed as still “at risk” beyond Day 300 in the voclosporin arm and the placebo arm, respectively.

Significantly greater reductions from baseline in UPCR were achieved in the voclosporin arm compared with the placebo arm at every time point.

Figure B.2-5. AURORA 1: Probability of ≥ 50% Reduction from Baseline in UPCR

==> picture [447 x 330] intentionally omitted <==

Abbreviation: HR = hazard ratio; UPCR = urine protein creatinine ratio.

Percentiles are Kaplan-Meier estimates. The HRs are from a Cox’s proportional hazards model with covariates for treatment group, baseline UPCR, biopsy class, mycophenolate mofetil use at baseline, and region Source: Rovin et al., 2021[2]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Disease activity (secondary endpoint)

Changes from baseline in disease activity were measured using the SELENA-SLEDAI instrument.[109] The SELENA-SLEDAI instrument objectively measures disease activity within the past 10 days by scoring 24 different disease activity descriptors.[116] Higher scores indicate a greater degree of disease activity, and the maximum theoretical score is 105 (all predictors are present). Improvements (i.e., decreases from baseline) in mean SELENA-SLEDAI index scores were observed in both treatment groups. Although numerically greater decreases from baseline were seen with voclosporin, there was no statistically significant difference between voclosporin and placebo (Table B.2-21).

Table B.2-21. AURORA 1: Change in SELENA-SLEDAI Index Score from baseline

Abbreviations: CI = confidence interval; SELENA-SLEDAI = Safety of Estrogens in Lupus Erythematosus National Assessment – Systemic Lupus Erythematosus Disease Activity Index Source: Otsuka 2020[109]

==> picture [44 x 10] intentionally omitted <==

Patient-reported outcomes

Improvements (i.e., increases) in mean scores from baseline were seen in both the voclosporin and the placebo arm for the HRQoL assessments SF-36 and for the healthrelated domains of the LupusPRO assessment.[109] Smaller changes were seen in both arms for the non-health-related domains of the LupusPRO assessment. There was no significant difference in the degree of improvement between the two treatments.

B.2.6.2 AURORA 2 Phase 3 long-term continuation study

AURORA 2 assessed the long-term safety, tolerability and efficacy of voclosporin compared with placebo for an additional 24 months following completion of treatment in the AURORA 1 study. A total of 216 eligible patients were analysed according to the treatment they were randomised to in the AURORA 1 study (n=100 in the placebo arm; n=116 in the voclosporin arm).[113]

AURORA 2 results reflected similar findings to AURORA 1, demonstrating favourable efficacy of voclosporin compared with placebo and a tolerable safety profile.[113]

==> picture [43 x 10] intentionally omitted <==

Complete renal response (secondary endpoint)

At months 18, 24, 30 and 36 during AURORA 2, the proportion of patients achieving CRR was higher in the voclosporin arm compared with the placebo arm (Table B.2-22).[113] Despite the study not being powered to measure statistical significance, a significant and clinically meaningful difference (p<0.05) from placebo was observed at every time point except the 36 month assessment (p=0.051).[113] In particular, voclosporin demonstrated significantly greater CRR than placebo at month 18 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx), month 24 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=0.035), and month 30 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx).[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-22. CRR at months 18 to 36

==> picture [44 x 9] intentionally omitted <==

Partial renal response (secondary endpoint)

A greater proportion of patients in the voclosporin arm also experienced PRR compared with patients in the placebo arm across all time points (defined as a 50% reduction from baseline in UPCR) (Table B.2-23).[113] As observed in AURORA 1, the high response rate in the placebo arm demonstrates that the MMF and steroid regimen used in the study is effective in reducing UPCR.[113] Despite this, voclosporin demonstrated significantly greater PRR than placebo at month 18 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx), month 24 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p= xxxx), and month 30 (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx).[113]

Table B.2-23. PRR at months 18 to 36

==> picture [44 x 9] intentionally omitted <==

Renal flares (secondary endpoint)

In order to be considered to have experienced a renal flare, patients must first achieve an adequate renal response.[113] Over the three-year course of the study, a greater proportion of patients in the voclosporin arm were considered to have an adequate renal response than those in the placebo arm (xxxx% versus xxxx%, respectively).[113] Among these patients, a slightly lower proportion of patients experienced a renal flare in the voclosporin arm compared to the placebo arm (xxxx% vs xxxx%, respectively) (Table B.2-24). Although a statistically significant difference could not be demonstrated between treatment arms, this may in part be due to the fact that AURORA 2 was not powered to demonstrate a significant difference in renal flare rates and few renal flare events were observed in either arm over the three year treatment period.[113]

When analysed on a year-by-year basis throughout the study period, the greatest difference in renal flare rate was observed during the first year of treatment; with fewer patients experiencing renal flares in the voclosporin arm compared with the placebo arm (xxx% vs xxxx%, respectively; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx).[113] In years two and three of

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 66 of 165

treatment, renal flares were similar between the voclosporin and placebo arms (Table B.2-24).[113]

Table B.2-24. Patients with adequate response and renal flares over the three year AURORA 1 and AURORA 2 study period

*A CEC adjudicated the response status of each patient, percentages for patients who responded are based on AURORA 2 population; percentages for patients with renal flares are based on the number of patients who responded prior to visit.

Abbreviations: CEC = Clinical Endpoints Committee; CI = confidence interval; OR = odds ratio Source: AURORA 2 CSR[113]

Due to the low number of patients with renal flares, a further analysis was conducted to assess and identify patients with sustained ‘good renal outcomes’ (i.e. those who achieved adequate response and did not experience renal flare). Significantly more patients in the voclosporin arm benefited from a good renal outcome than those in the placebo arm (xxxx% vs xxxx%; OR xxxx [95% CI xxxxxxxxx]; p=xxxxx), demonstrating a clear-long-term renal benefit of voclosporin treatment (Table B.2-25).[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-25. Patients with good renal outcomes over the three-year AURORA 1 and AURORA 2 study period*

*Good renal outcome is defined as adequate response and without flare. Abbreviations: CI = confidence interval; OR = odds ratio Source: AURORA 2 CSR[113]

==> picture [44 x 10] intentionally omitted <==

Extra-renal flares (secondary endpoint)

Independent of renal response status, patients could experience non-renal (“extra-renal”) flares at any point during the AURORA trials.[113] During the three-year study period, xxxx% of patients in the placebo arm and xxxx% of patients in the voclosporin arm were considered to have extra-renal flares (OR xxxx [95% CI xxxx–xxxx; p=xxxxx) (Table B.2-26). As with other efficacy endpoints, AURORA 2 was not powered to demonstrate a significant difference in extra-renal flares and there were notably few occurrences of extra-renal flare in the AURORA 2 study population (as is typically the case in patients with LN).[113]

Table B.2-26. Patients with extra-renal flares over the three-year AURORA 1 and AURORA 2 study period

==> picture [44 x 9] intentionally omitted <==

Disease activity (secondary endpoint)

Disease activity, as measured via the SELENA-SLEDAI score (see Section B.2.6.1.6), was higher in the voclosporin arm (mean: xxxx, median: xxxx) than the placebo arm (mean: xxxx, median: xxxx).[113] Improvements from AURORA 1 baseline were seen in both arms during AURORA 2, demonstrating improvements in SLE symptoms. The greatest improvements were observed during the first year of treatment, however there was no significant difference between the treatment arms.[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Change in UPCR from baseline

At the start of AURORA 1, baseline mean UPCR levels were balanced between the two treatment arms (xxxx mg/mg in the placebo arm and xxxx mg/mg in the voclosporin arm).[113] At the end of AURORA 1 (Month 12), mean UPCR was xxxx in the voclosporin arm compared with xxxx in the placebo arm. At the follow-up visit, UPCR in the voclosporin arm showed a decrease of xxxx from baseline compared with a decrease of xxxx in the placebo arm.[113]

The MMRM analysis confirmed that statistically significantly greater reductions from baseline in UPCR were achieved in the voclosporin arm compared with the placebo arm at Months 18, 24 and 30 but not Month 36.[113]

During AURORA 2 (from Month 12) there was little change in UPCR in either treatment arm. Mean UPCR values at Month 12 were lower in the voclosporin arm (xxxx mg/mg) than in the placebo arm (xxxx mg/mg) as a result of benefit derived from 12 months of treatment with voclosporin.[113] There was no demonstrable difference between the two arms in the change from Month 12 at visits through to Month 36, showing that the difference observed at Month 12 is sustained for a further 2 years with continued treatment with voclosporin.[113]

==> picture [44 x 10] intentionally omitted <==

Change in urine protein, serum creatinine and eGFR from baseline

Urine protein decreased across the 3 years of observation during the AURORA 1 and AURORA 2 studies.[113] There was a greater decrease in mean urine protein observed in patients receiving voclosporin compared with placebo, which was consistent with UPCR findings.[113] The MMRM analysis confirmed a statistically significantly greater mean decrease for voclosporin treatment compared to placebo at most time points.[113]

Mean serum creatinine levels at baseline prior to the start of treatment in AURORA 1 were within normal range and similar in both treatment arms (placebo: xxxxx mg/dL, voclosporin: xxxxx mg/dL).[113] Over the first 15 months of treatment, small increases (i.e. within normal range) in mean levels were observed in the voclosporin arm while levels in the placebo arm decreased slightly.[113] This resulted in statistically significant differences between the treatment arms up to Month 15 in the MMRM analysis but not from Month 18 onwards.[113] During AURORA 2, mean corrected eGFR values were similar in both arms prior to the start of study treatment in AURORA 1 (xxxx mL/min/1.73m[2] in the voclosporin arm and xxxx mL/min/1.73m[2] in the placebo arm).[113] Over the first 3 months of treatment, the mean corrected eGFR were stable in the voclosporin arm while the mean value in the placebo arm showed a small increase. The xxxxxxxxxxxxxxxx between the arms remained through to Month 27, after which the mean eGFR value increased slightly in the voclosporin arm and started to decline in the placebo arm.[113]

==> picture [44 x 10] intentionally omitted <==

Patient reported outcomes

Improvements (i.e. increases) in mean scores from baseline were seen in both the voclosporin and the placebo arm for all domains of the SF-36 assessment, with no significant difference in the total mean scores observed between the two treatments.[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.2.6.3 AURA-LV Phase 2 study

AURA-LV was a Phase 2, 48-week, randomised, double-blind, parallel-group, placebocontrolled, three-arm, multicentre study designed to compare the efficacy and safety of two doses (high- and low-dose) of voclosporin and placebo in patients with LN.

The study met its primary objective, demonstrating a higher proportion of patients achieved CRR after 24 weeks in the voclosporin groups than in the placebo group.

==> picture [43 x 9] intentionally omitted <==

Complete Renal Response at Week 24 (primary endpoint)

At Week 24, CRR was achieved by a higher proportion of patients in both the low-dose (32.6%) and high-dose (27.3%) voclosporin groups compared to the placebo group (19.3%). CRR at Week 24 was significantly improved in patients treated with low-dose voclosporin compared to patients in the placebo group (OR=2.03; [95% CI: 1.01, 4.05]; p=0.045).[8]

==> picture [44 x 10] intentionally omitted <==

Complete Renal Response at Week 48 (secondary endpoint)

At Week 48, CRR was achieved by a higher proportion of patients in both the low-dose (49.4%) and high-dose (39.8%) voclosporin groups compared to the placebo group (23.9%), with an increased separation between the treatment and control arms compared to Week 24. CRR was increased in both the voclosporin groups compared to the placebo: i.e., patients treated with low-dose voclosporin had triple the odds of achieving CRR at Week 48 compared to patients in the placebo group (OR=3.21; [95% CI: 1.68, 6.13]; p<0.001), and patients treated with high-dose voclosporin had double the odds of achieving CRR compared to patients in the placebo group (OR=2.10; [95% CI: 1.09, 4.02]; p=0.026).[8]

==> picture [44 x 10] intentionally omitted <==

Partial renal response at Week 24 and Week 48 (secondary endpoint)

At Week 24, partial renal response was achieved by a higher proportion of patients in both the low-dose (69.7%) and high-dose (65.9%) voclosporin groups compared to the placebo group (49.4%).[114] Low-dose or high-dose voclosporin had double the odds of achieving partial renal response at Week 24 compared to patients in the placebo group (OR 2.33; p=0.007 and OR=2.03; p=0.024, respectively). Results were similar at Week 48, with even higher odds demonstrated for the high-dose voclosporin group versus placebo (OR 2.68; p=0.002).[114]

==> picture [44 x 9] intentionally omitted <==

Time to Complete Renal Response (secondary endpoint)

CRR occurred statistically significantly earlier in patients treated with either low-dose or highdose voclosporin compared to placebo (HR 2.26 and 2.25, respectively). The median time to CRR was 19.7 weeks in the low-dose voclosporin group and 23.4 weeks in the high-dose voclosporin group. Median time to CRR could not be determined for the placebo group (Figure B.2-6).[114]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 70 of 165

Figure B.2-6: AURA-LV: Analysis of Time to CRR

==> picture [421 x 266] intentionally omitted <==

Source: Otsuka 2018[114]

==> picture [44 x 9] intentionally omitted <==

Time to Partial Renal Response, Sustained Partial Renal Response and Sustained Early Partial Renal Response (secondary endpoint)

Partial renal response occurred significantly earlier in patients treated with either low-dose or high-dose voclosporin compared to placebo (HR 1.63 (p=0.005) and HR 1.74 (p=0.002), respectively). The median time to partial renal response was 4.3 and 4.4 weeks in the lowdose and high-dose voclosporin groups, respectively, compared to 6.6 weeks in the placebo group.[8,114]

Compared to placebo, sustained partial renal response occurred significantly earlier in patients treated with either low-dose voclosporin (HR=2.03; p<0.001) or high-dose voclosporin (HR=1.81; p=0.004).[114] The median time to sustained partial renal response was 26.9 weeks in the placebo group, compared to 6.3 weeks in the low-dose voclosporin group and 8.1 weeks in the high-dose voclosporin group.[114]

Sustained early partial renal response was achieved by a higher proportion of patients in both the low-dose (67.4%) and high-dose (65.9%) voclosporin groups compared to the placebo group (41.4%).[114] Both voclosporin dose groups demonstrated that significantly increased odds of achieving sustained early partial renal response compared to patients in the placebo group. The patients treated with low-dose voclosporin had an OR of 2.93 compared to those treated with placebo (p<0.001) and the patients treated with high-dose voclosporin had an OR of 2.74 compared to those treated with placebo (p=0.021).[114]

Compared to placebo, time to sustained early partial renal response occurred significantly earlier in patients treated with either low-dose voclosporin (HR=2.21; p<0.001) or high-dose voclosporin (HR=1.87; p=0.004).[114] The median time to sustained early partial renal response was 6.3 weeks in the low-dose voclosporin group and 8.1 weeks in the high-dose voclosporin group. Median time to CRR could not be determined for the placebo group.[114]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [44 x 9] intentionally omitted <==

Disease activity

Mean SELENA-SLEDAI scores improved (i.e., decreased) in all 3 treatment groups. Changes from baseline in mean SELENA-SLEDAI scores were significantly greater for both the lowdose and high-dose voclosporin groups compared with placebo at Week 24 (p=0.003 for both comparisons) and at Week 48 (p<0.001 for both comparisons; Table B.2-27).[114]

Table B.2-27. AURA-LV: Mean Change from Baseline in SELENA-SLEDAI Scores at Week 24 and Week 48

*23.7 mg BID; †39.5 mg BID; ‡Significant difference compared with placebo ( p <0.05) in ANCOVA for the change from baseline

Abbreviations: ANCOVA = analysis of covariance; CI = confidence interval; SELENA-SLEDAI = Safety of Estrogens in Lupus Erythematosus National Assessment - Systemic Lupus Erythematosus Disease Activity IndexNote: a decrease in SELENA-SLEDAI score indicates improvement Source: Otsuka 2018[114]

B.2.7 Subgroup analysis

B.2.7.1 AURORA 1 Phase 3 study

==> picture [43 x 9] intentionally omitted <==

Methodology and statistical analysis

The primary endpoint of CRR at Week 52 was analysed for the following pre-specified subgroups:[109]

• Age (≤30 vs >30 years)

• Gender (male, female)

• Race (White, Asian, other)

• Biopsy class (class V, other)

• Region (Asia-Pacific, Europe and South Africa, Latin America, North America)

• MMF use at screening (yes, no)

• Maximum MMF dose (≤2 g vs >2 g)

Prespecified covariate analyses were done using a logistic regression model. An interaction between the subgroup and treatment group was added to the model, and a p value for the main effect of the covariate in question along with the p-value for the interaction between treatment and covariate were reported.[109]

==> picture [44 x 9] intentionally omitted <==

Results of subgroup analyses

Results of the subgroup analyses are presented in Appendix E. The treatment benefit of voclosporin was seen in all pre-specified subgroups.[109] Although the study was not powered to detect a significant difference between the two treatments in the individual subgroups, statistically significant results were observed for many subgroups, confirming the positive effect of voclosporin in achieving renal response. Where the results were not statistically significant (White, pure Class V, Europe + South Africa, North America, no MMF at screening and maximum MMF dose >2 g), the odds ratios still favoured voclosporin over placebo.[109] Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

B.2.7.2 AURORA 2 Phase 3 long-term continuation study

Subgroup analyses were not planned for the AURORA 2 study, nor have any post-hoc analyses been conducted at the time of this submission.

B.2.7.3 AURA-LV Phase 2 study

==> picture [43 x 9] intentionally omitted <==

Methodology and statistical analysis

No subgroup analyses were stipulated in the Statistical Analysis Plan (SAP). However, posthoc subgroup analyses were conducted for CRR at Weeks 24 and 48 to explore the impact of the imbalance in randomisation of low gross domestic product (GDP) patients (i.e. low-GDP and non-low GDP) and biopsy class (i.e. class III, III/V, IV, IV/V, and V). TEAEs and serious TEAEs were analysed according to GDP subgroups.[114]

Covariate analyses were also conducted for CRR at Week 24 and 48 including the following covariates:[114]

• Age (≤30 vs >30 years)

• Gender (male, female)

• Race (White, Asian, other)

• Biopsy class (class V, other)

• Region (Asia-Pacific, Europe and South Africa, Latin America, North America)

• MMF use at screening (yes, no)

• Maximum MMF dose (≤2 g vs >2 g)

==> picture [44 x 10] intentionally omitted <==

Results of subgroup analyses

Results of the subgroup and covariate analyses are presented in Appendix E.[114]

B.2.7.3.2.1 Subgroup analysis: CRR and TEAES/serious TEAEs in GDP subgroups

At Week 24, the CRR rate was notably lower for both voclosporin dose groups within the lowGDP subgroup, particularly for those treated with high-dose voclosporin (low GDP: 12.1% vs non-low GDP: 36.4%). The impact was less pronounced at Week 48, with little difference in CRR between the overall population or GDP subgroups.[114] Across both voclosporin dose groups, CRR rates at Week 24 increased when low-GDP patients were excluded (i.e. from 32.6% to 38.3% in the low-dose group and from 27.3% to 36.4% in the high-dose group).[114]

When low-GDP patients were excluded, the overall incidence of TEAEs was also reduced, especially in the two voclosporin groups. In addition, a similar incidence of serious TEAEs and TEAEs leading to death was observed in patients in non-low GDP countries among all three treatment groups.[114]

B.2.7.3.2.2 Subgroup analysis: CRR in biopsy subgroups

At both Week 24 and Week 48, a trend favouring low-dose voclosporin over placebo was maintained across all biopsy classes apart from pure class V. The results for an “all but pure class V” subgroup were consistent with the results for the overall population.[114]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 73 of 165

B.2.7.3.2.3 Covariate analyses

Low-dose voclosporin had a beneficial effect in terms of CRR at Week 24 across most covariates compared to placebo. The treatment benefit was not statistically significant for the majority of strata; however, this was likely due to the small sample size (e.g. male gender (n=28) and “other” race (n=17)).[114] ORs in favour of low-dose voclosporin were statistically significant for female gender; “other” biopsy class (i.e. not Class V); no MMF use at screening; White race; the region of Europe; and age >30 years. Odds ratios favoured placebo for male gender (OR 0.30) and Class V biopsy class (OR 0.19), although the results were not statistically significant (p=0.206 and p=0.075, respectively). Overall, similar trends were seen in the covariate analysis for the comparison of high-dose voclosporin versus placebo.[114]

B.2.8 Meta-analysis/pooled analysis

An integrated analysis of AURORA 1 and AURA-LV has been completed to provide more information on the treatment effect for voclosporin. A pooled LN population was defined to comprise patients exposed to voclosporin (23.7 mg twice daily; n=268) or placebo (n=266), each in combination with MMF and low-dose corticosteroids per AURORA 1 and AURA-LV dosing regimens for up to one year.[110]

Within the pooled dataset, key demographics and baseline characteristics were comparable between treatment arms (median age: 30 vs 32; proportion of Hispanic or Latino: 25% vs 27%; median eGFR: 92 vs 98 ml/min/1.73 m[2] ; median UPCR: 3.5 vs 3.1 mg/mg; median time since LN diagnosis: 2.2 vs 2.2 years for voclosporin and placebo arms, respectively), including an identical proportion of patients with class III or IV± V (38%) or pure class V (14%) LN.[110]

CRR was analysed using a logistic regression model that included terms for study, treatment group, baseline UPCR, biopsy class, MMF use at screening, and region with adjudicated renal response outcomes at one year as the response variable.[117] Time to UPCR ≤0.5mg/mg and time to ≥50% reduction in UPCR were estimated using Kaplan-Meier methodology, with a Cox proportional hazards model fitted using terms for study, baseline UPCR, biopsy class, MMF use at screening, and region.[117] Change from baseline in UPCR was also analysed using a MMRM analysis with study, treatment arm, visit, treatment by visit interaction, treatment by study interaction, biopsy class, MMF use at screening, region, and baseline parameter (UPCR/serum creatinine/eGFR) included as covariates in the model.[117]

In summary, CRR rates were significantly greater in the voclosporin arm compared to the placebo arm at both six months (31.7% vs 20.3%, respectively; OR: 2.01, p=0.008) and one year (43.7% vs 23.3%, respectively; OR: 2.76, p<0.0001).[110] Similarly, a significantly greater proportion of patients achieved PRR in the voclosporin arm at both six months (70.1% vs 49.8%; OR: 2.42; p=<0.0001) and one year (69.4% vs 50.6%; OR: 2.26; p<0.0001) compared to placebo.[117] A ≥50% UPCR reduction was also achieved in 93.7% of patients in the voclosporin arm, and 75.2% of patients in the control arm; and the median time to ≥50% UPCR reduction was significantly shorter for voclosporin relative to placebo (29 days vs 58 days, respectively; HR: 1.96, p<0.0001).[110] Decreases in mean UPCR were observed at Week 4 and sustained over a 52 week period for both treatment arms, with a significantly greater reduction from baseline observed in the voclosporin arm compared to the placebo arm at Week 52 (mean UPCR -1.1 mg/mg; p<0.0001).[117]

Apart from a head-to-head data for voclosporin versus MMF provided by AURORA 1, AURALV, and AURORA 2, other head-to-head evidence is not available to compare voclosporin to

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

alternative comparators (i.e. rituximab, cyclophosphamide, tacrolimus, tacrolimus+MMF and azathioprine). Therefore, a network meta-analysis (NMA) was performed to estimate the relative efficacy of voclosporin versus all relevant comparators (Section B.2.9).

B.2.9 Indirect and mixed treatment comparisons

In the absence of head-to-head data, an NMA was conducted to compare the efficacy of voclosporin to all relevant comparators using published evidence identified from the clinical SLR. NMA efficacy outcomes of interest were pre-defined as CRR and PRR, and NMA results were used to inform comparator short-term efficacy in the cost-effectiveness model (Section B.3). As CRR and PRR are mutually exclusive health states by definition, the PRR network only included patients who achieved a PRR, independent of CRR. Therefore, trials were removed from the NMA if they did not report PRR independently of CRR.

B.2.9.1 Search strategy and study selection for the network meta-analysis

A full overview of the SLR methods undertaken for this submission are provided in Appendix D. Systematic searches were conducted on 1[st] June 2021, and later repeated on 24 January 2022 to identify RCTs that evaluated the efficacy and safety of active treatments in patients with active LN. A total of 57 publications reporting on 44 unique trials were identified from the databases. To present and describe the key evidence relevant to the final scope, networks that were dependent on the outcomes of interest were constructed by selecting only those RCTs that evaluated voclosporin and the comparators of interest for the treatment of patients with active LN. For a full overview of the trials included and excluded from the NMA, refer to Appendix D.

Criteria for study inclusion/exclusion to define the NMA evidence base are described in Table B.2-28.

Table B.2-28. Criteria for study inclusion/exclusion used in selection for the NMA evidence base

Abbreviations: AD = active disease; AZA = azathioprine; BID = twice daily; CNI = calcineurin inhibitor; CRR = complete renal response; CYC = cyclophosphamide; MMF = mycophenolate mofetil; PRR = partial renal response; RCTs = randomised controlled trials; RTX = rituximab

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 75 of 165

B.2.9.2 Summary of trials included in the NMA

A summary of RCTs included in the base case NMA and in the scenario analyses are described in Appendix D for the pre-defined outcomes of CRR and PRR, respectively. In addition, Appendix D contains an overview of CRR and PRR outcomes for studies included in the networks for each comparator, with outcomes presented at ≤6 month and ≥12 month follow-up where possible.

The base case treatment network for the CRR outcome is presented in Figure B.2-7, and includes a total of 17 RCTs reporting on 8 treatments in an overall patient population. Scenario analysis networks are also presented in Appendix D, to include trials that are non-essential (i.e. those that contribute additional evidence to the network but do not provide essential links), to exclude trials with a substantially different outcome definition of CRR or those with a 100% Asian patient population, and to assess response at different lengths of follow-up.

Figure B.2-7. Treatment network for studies contributing to evidence for CRR in the overall population

==> picture [379 x 242] intentionally omitted <==

Abbreviations: AZA = azathioprine; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

The base case treatment network for PRR outcome is presented in Figure B.2-8, and includes a total of 10 RCTs reporting on 6 treatments in an overall patient population. Although Zhang 2020 reported on PRR for tacrolimus + MMF, it could not be connected to the PRR network as no other trials in this network included tacrolimus + MMF or MMF + cyclophosphamide as comparators. Scenario analyses consisted of the same scenarios as performed for CRR, the networks for each scenario are presented in Appendix D.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 76 of 165

Figure B.2-8. Treatment network for studies contributing to evidence for PRR in the overall population

==> picture [343 x 252] intentionally omitted <==

Abbreviations: H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

B.2.9.3 Heterogeneity assessment of trials included

Study similarity was assessed for heterogeneity according to the population, intervention, comparator, outcome, study design (PICOS) framework detailed in Table B.2-29. Tables with baseline patient characteristics, summary of outcome definitions and background corticosteroid use are provided in Appendix D.

Table B.2-29. PICOS items for heterogeneity assessment

Abbreviations: CRR = complete renal response; MMF = mycophenolate mofetil; PRR = partial renal response

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 77 of 165

Trial designs and patient population across included studies were generally comparable (Appendix D), although the following key differences were observed:

• Doria 1994 (included in scenario analyses) assessed the efficacy and safety of azathioprine combined with corticosteroids, whereas other studies reported corticosteroid use as background therapy

• Most included studies did not report the study phase, with only the AURORA 1 and LUNAR trials listed as Phase 3. Other studies tended to include a smaller sample size

• The dosage of MMF (the NMA reference treatment) varied between included studies, while AURORA-1 and AURA-LV used a dosage which was lower than in some studies (1 mg)

• Trial length of follow-up varied between studies, with most reporting outcomes at six months only

• Outcome definitions for CRR and PRR varied across studies. Some studies required patients to fulfil many criteria to achieve renal response, while others included less stringent criteria

• Yap 2012 and Doria 1994 were randomised controlled pilot studies. Therefore, few patients were assigned to treatment (<10 patients per treatment arm in each study). Wang 2007 was also a small sample study (phase not reported), and included only 20 patients across two treatment arms

• Six of the included trials were restricted to the Asian-Pacific region and consisted of Asian patients only (Feng 2014, Kamanamol, Li 2012, Liu 2015, Mok 2016, and Yap 2012)

B.2.9.4 Statistical methods for the network meta-analysis

The NMA was conducted in a Bayesian framework using Monte Carlo Markov Chain (MCMC) and implemented using models developed in the probabilistic modelling language of Stan (Version 2.21.0).[118] A generalised linear model for dichotomous outcomes was applied, as presented within the NICE Decision Support Unit (DSU) Technical Support Document (TSD) 2.[119] Treatment effects were synthesised using the observed number of events from the known number of patients in the respective treatment arms. The data was assumed to come from a binomial likelihood. Therefore, the binomial model with a logit link was used to model the log odds of the outcome on a given treatment, in a specified trial via an effect (fixed or random). As recommended, a non-informative prior was assigned for the treatment effect, in both fixed and random effects models, N(0, 100[2] ). A more informative prior, half~N(0, 5), was applied for the parameter representing the between study variation in the random effects model.

In the Monte Carlo simulation, 4 simulation chains with a minimum of 10,000 iterations (including 5,000 burn-in) were used to summarise the posterior distribution. The number of samples was deemed appropriate for model convergence. Convergence was then assessed in accordance with NICE DUS TSD 2; by examining diagnostic autocorrelation, trace, and density plots as well as the recommended statistics such as the Gelman-Ruben Rhat, and whether the Monte Carlo standard errors are ≤5% of the posterior deviation of the parameters of interest.[119]

Evidence networks were also assessed for inconsistency between the direct and indirect evidence, by comparison of the standard consistency model and with an inconsistency model, as proposed in the NICE DSU TSD 4.[120] Deviance contributions from each fitted point were assessed along with the effective number of parameters, and deviance information criterion (DIC) were compared to assess model fit and validity of the consistency assumption.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 78 of 165

Detailed results and plots for the consistency checks are provided in Appendix D for the outcomes of CRR and PRR. Potential inconsistency was discovered in the CRR network, arising from the Feng 2014 study. A careful review of the evidence was undertaken, and no data extraction errors were identified. A comparison between the unrestricted means (UME) inconsistency model and the standard consistency model, for the fixed effects NMA presented, returned no significant differences between model fit and DIC (Appendix D). Comparisons for the UME model and consistency model for the PRR demonstrated equivalence between the models in terms of the deviance contributions and the consistency model is deemed appropriate.

Full details of the statistical methods adopted for the NMA are provided in Appendix D.

B.2.9.5 Results of the network meta-analysis

NMA results of relevance to the decision problem are summarised in Table B.2-30 for CRR and Table B.2-32 for PRR. For indirect comparisons, MMF was selected as the reference treatment for which all other treatments are compared to, as MMF was the most common comparator across trials. Pairwise ORs of all treatment comparisons are provided in Table B.2-31 and Table B.2-33 for CRR and PRR, respectively. Results of additional scenario analyses not of relevance to the decision problem have been reported in Appendix D.

For the CRR outcome, the NMA estimated a high probability (≥95%) for voclosporin + MMF to be more efficacious than MMF in the overall population, thereby reiterating the results of the AURA-LV and AURORA-1 clinical trials for voclosporin + MMF (median OR xxxx [95% credible interval (CrI):xxxxxxxxx]; Figure B.2-9). No further treatments demonstrated a greater efficacy than MMF in terms of CRR, and both the high-dose (H-CYC) and low-dose (L-CYC) regimens for cyclophosphamide were inferior to MMF in terms of achieving a CRR (median OR xxxx [95% CrI: xxxxxxxxx] and xxxx [95% CrI: xxxxxxxxx], respectively).

Surface under the cumulative ranking curve (SUCRA) values are provided for the ranking of treatments in Table B.2-30 and Table B.2-32 for CRR and PRR, respectively. The SUCRA shows that voclosporin + MMF is highly likely to be the preferred treatment option with a value of xx%, followed by the combination therapy of tacrolimus + MMF (xx%) and the reference MMF (xx%).

Table B.2-30. Primary analysis, fixed-effects network meta-analysis for CRR

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Figure B.2-9. Forest plot for posterior median ORs and 95% CrI, for CRR

==> picture [452 x 352] intentionally omitted <==

Abbreviations: AZA = azathioprine; CrI = credible interval; CRR = complete renal response; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; OR = odds ratio; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 80 of 165

Table B.2-31. Pairwise odds ratios for CRR, OR (95% CrI)

Abbreviations: AZA = azathioprine; CrI = credible Interval; CRR = complete renal response; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MPR = methylprednisolone; MMF = mycophenolate mofetil; OR = odds ratio; PR = prednisolone; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 81 of 165

For the PRR outcome, the NMA indicated that voclosporin + MMF and rituximab + MMF have a high probability (≥95%) of being more efficacious than MMF in the overall population based on studies that reported partial responders independently from those who achieved a CRR (median OR xxxx [95% CrI: xxxxxxxxx] and xxxx [95% CrI: xxxxxxxxx], respectively; Figure B.2-10). On the other hand, neither cyclophosphamide regimens nor tacrolimus were significantly different to MMF in achieving PRR. Furthermore, the SUCRA demonstrated that voclosporin + MMF was the second most likely regimen to be the preferred treatment option when considering an independent PRR (SUCRA: xx%), behind rituximab + MMF (xx%) but ahead of tacrolimus (xx%), and the reference MMF (xx%).

Table B.2-32. Primary analysis: fixed-effects network meta-analysis for PRR

Notes: Values underlined demonstrate a high probability (≥95%) of being more efficacious than MMF Abbreviations: CrI = credible interval; DIC = deviance information criterion; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; OR = odds ratio; pD = parameters; PRR = partial renal response; RTX = rituximab; SUCRA = surface under the cumulative ranking curve; TAC = tacrolimus; VCS = voclosporin

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Figure B.2-10. Forest plot for posterior median ORs and 95% CrI, for PRR

==> picture [452 x 303] intentionally omitted <==

Abbreviations: AZA = azathioprine; CrI = credible interval; CRR = complete renal response; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; OR = odds ratio; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

Table B.2-33. Pairwise odds ratios for PRR, OR (95% CrI)

Abbreviations: AZA = azathioprine; CrI = credible Interval; CRR = complete renal response; H-CYC = high-dose cyclophosphamide; L-CYC = low-dose cyclophosphamide; MMF = mycophenolate mofetil; OR = odds ratio; PR = prednisolone; RTX = rituximab; TAC = tacrolimus; VCS = voclosporin

B.2.9.6 Uncertainties in the indirect and mixed treatment comparisons

While a NMA allows for the indirect comparison of voclosporin + MMF versus the comparators relevant to the decision problem, some uncertainties exist within the approach taken.

The binomial approach considered does present some limitations, as the included trials do not all have the same follow-up time. The logit model assumes either that all patients who reach

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

the endpoint, do so by a specific follow-up time, and further follow-up makes no difference; or that the proportional odds assumption holds.[119] However, scenario analyses at different time points indicated that the ORs changed at the 6-month analysis to that of the >1 year analysis (the base case used the longest-follow up available up to a maximum of 2-years). The length of follow-up available was also identified as a contributor for heterogeneity of the studies included, as many only reported on outcomes for 6-months.

A further limitation was the heterogeneity observed between the studies included. This was present for several factors and one that would considerably affect the number of events was the definition of response across trials. For example, the AURORA 1 and AURA-LV trials required a more stringent definition of CRR as patients were required to achieve an additional eGFR component, whereas in other trials, CRR was determined if the proteinuria or UPCR condition was met. In addition, there was some imbalance in patient characteristics across trials, mainly in terms of patient race. Although a scenario analysis was conducted to remove studies with a 100% Asian patient population, this inadvertently led to the removal of all evidence for the comparator of tacrolimus alone. Thus, the only evidence that contributed for tacrolimus was from trials that only contained Asian patients. Further to this, there was a lack of reporting on subgroups; with few trials reporting outcomes for subgroups of interest (i.e. those presented in the AURA-LV and AURORA-1 studies), so networks were unable to be constructed for analysis.

Finally, the differences between the studies included are likely due to largely off-label treatments being used in the treatment of LN so, historically, there has been a lack of highquality pivotal studies designed and developed for regulatory HTA purposes. A resulting outcome of this is that many of the trials included were relatively small in sample size, mostly with less than 50 patients in the entire study, and therefore contributes to the uncertainty of the estimates from the NMA.

B.2.9.7 Conclusions

In summary, the results of the NMA indicate that treatment of active LN with voclosporin + MMF is superior to current standard of care immunosuppressant treatments for inducing a CRR (OR xxxx [95% CrI: xxxxxxxxx]) and highly likely to be ranked the best treatment in terms of SUCRA. The primary analysis also showed voclosporin + MMF to be similar to current treatments in inducing a PRR; although this is likely due to the large number of patients in AURORA 1 and AURA-LV achieving a CRR over a PRR.

B.2.10 Adverse reactions

==> picture [49 x 10] intentionally omitted <==

Safety population

This submission is supported by safety data from a Phase 3 study (AURORA 1), a long-term Phase 3 continuation study (AURORA 2), and the Phase 2 study (AURA-LV). A summary of patients evaluable for safety and toxicity in each study is presented in Table B.2-34.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 84 of 165

Table B.2-34. Safety populations

*In addition to oral corticosteroid and MMF

Abbreviations: BID = twice daily; MMF = mycophenolate mofetil; N/n = number of patients evaluable; N/A = not applicable Sources: Otsuka 2018, 2020 and 2022[109,113,114]

==> picture [50 x 10] intentionally omitted <==

Extent of exposure

The extent of exposure to voclosporin for each study (AURORA 1, AURORA 2, and AURALV) is presented in Table B.2-35.

Table B.2-35. Extent of exposure to voclosporin

**In addition to oral corticosteroid and MMF

Abbreviations: mg = milligram; MMF = mycophenolate mofetil; n = number of patients evaluable; NR = not reported Source: Otsuka 2018, 2020 and 2022[109,113,114]

B.2.10.2 AURORA 1 Phase 3 study

==> picture [49 x 9] intentionally omitted <==

Adverse events

Overall and serious AEs occurred at similar frequencies in both treatment groups, and most AEs were of mild or moderate intensity (Table B.2-36).[2] The most frequent type of AE in both groups was infections and infestations, which is expected in this immunocompromised patient population.[109]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 85 of 165

Table B.2-36 AURORA 1: summary of AEs

Abbreviations: AE = adverse event; n = number of patients; NR = not reported; TEAE = treatment-emergent adverse event Source: Otsuka 2020[109]

==> picture [50 x 9] intentionally omitted <==

Commonly reported adverse events

Approximately 90% of patients in both arms experienced at least one TEAE (voclosporin arm: 162 [91.0%]; placebo arm: 158 [88.8%]. The most common TEAEs in both groups were Infections and Infestations, reported by 64.6% of patients in the voclosporin arm and 56.7% of patients in the placebo arm (Table B.2-37). The most frequent infections in both arms were upper respiratory tract infections (URTIs) and urinary tract infections (UTIs). The majority of infections were of mild or moderate intensity; severe infections (predominantly pneumonia), were recorded in 10 patients (5.6%) in the voclosporin arm and 7 patients (3.9%) in the placebo arm.[109]

Known side effects of MMF use include diarrhoea, nausea, vomiting and dyspepsia. Gastrointestinal Disorders were the second most common TEAEs. More gastrointestinal events were recorded in patients in the voclosporin arm than the placebo arm (46.6% vs 34.3%), particularly diarrhoea and abdominal pain/upper abdominal pain.[109]

Known adverse effects of CNIs, such as diabetes, kidney dysfunction and hypertension, were also of particular interest in this study.[2,109] New onset diabetes did not occur in any voclosporintreated patients and in 1 placebo-treated patient,[2] the incidence of investigator-reported serious renal dysfunction was low and similar between treatment groups (voclosporin, 3%; placebo, 2%),[2] and overall, there was no significant difference in mean blood pressure between the treatment groups.[2]

Table B.2-37. AURORA 1: Most common TEAEs (occurring in ≥ 4% of patients in any group)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 86 of 165

Abbreviations: TEAE = treatment-emergent adverse event Source: Otsuka 2020;[109] Rovin et al., 2021[2]

Treatment-related TEAEs were reported in 44.9% and 25.3% of patients in the voclosporin and placebo arms, respectively. The majority of treatment-related TEAEs were of mild or moderate intensity, with severe events recorded in 12 patients (6.7%) in the voclosporin arm and two patients (1.1%) in the placebo arm. The most common treatment-related TEAE was GFR decreased (18.0% vs 2.8%, respectively).[109] Hemodynamically mediated decreases in GFR are known to be associated with CNIs and so this outcome was not unexpected. Vascular disorders (predominantly hypertension) and renal and urinary disorders were also considered treatment-related in a greater proportion of patients in the voclosporin arm than the placebo arm (hypertension: 7.3% vs 1.7%, respectively; renal and urinary disorders: 4.4% vs 1.7%, respectively).[109]

==> picture [50 x 9] intentionally omitted <==

Serious adverse events

A similar proportion of patients in each arm experienced serious TEAEs (voclosporin arm: 37 [20.8%]; placebo arm: 38 [21.3%]).[2,109] The most common serious TEAEs (reported in ≥2 patients in any treatment group) are summarised in Table B.2-38.

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-38. AURORA 1: Most common serious TEAEs (in ≥2 patients in any group)

Abbreviations: TEAE = treatment-emergent adverse event Source: Otsuka 2020[109]

Serious treatment-related TEAEs were observed in the same number of patients in each treatment group (voclosporin arm: 8 [4.5%]; placebo arm: 8 [4.5%]).[109] Serious treatmentrelated TEAEs are summarised in Table B.2-39.

Table B.2-39. AURORA 1: Serious treatment-related TEAEs

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

*including cysts and polyps Abbreviations: TEAE = treatment-emergent adverse event Source: Otsuka 2020[109]

==> picture [50 x 10] intentionally omitted <==

Deaths

Mortality was lower in the voclosporin group of this study (Table B.2-36). Three placebotreated patients died as a result of TEAEs (pneumonia; pneumonia and septic shock; LN). An additional two patients in the placebo group and one patient in the voclosporin group died due to AEs that started more than 30 days after the last dose of study drug. None of the events leading to death were considered by the investigators to be related to study treatment.[109]

==> picture [51 x 10] intentionally omitted <==

Adverse events leading to treatment discontinuation

A similar proportion of patients in the voclosporin and placebo arm had their study treatment discontinued as a result of a TEAE; 20 patients (11.2%) in the voclosporin arm and 26 patients (14.6%) in the placebo arm had their study drug discontinued as a result of a TEAE, most commonly this was due to Renal and Urinary Disorders.[109] A summary of most common TEAEs leading to treatment discontinuation is presented in Table B.2-40.

Table B.2-40. AURORA 1: Most common TEAEs leading to treatment discontinuation (in ≥2% of patients in any group)

Abbreviations: TEAE = treatment-emergent adverse event Source: Otsuka 2020[109]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

==> picture [50 x 9] intentionally omitted <==

Adverse events leading to dose interruption or modification

More patients in the voclosporin arm (80 patients [44.9%]) than the placebo arm (47 patients [26.4%]) had their dose of study drug modified as a result of a TEAE.[109]

As expected for a CNI, the most common TEAE leading to dose modification was GFR decreased (reported for 40 patients [22.5%] in the voclosporin arm and 11 patients [6.2%] in the placebo arm.[109] However, only 3 patients in the voclosporin arm and 4 in the placebo arm had their treatment permanently discontinued as a result of decreased GFR (Table B.2-40). Serious TEAEs resulting in study drug dose modifications were reported for 19 patients (10.7%) in the voclosporin arm and 15 patients (8.4%) in the placebo arm; these were predominantly infections (in 11 voclosporin patients [6.2%] and 10 placebo patients [5.6%]). A summary of most common TEAEs leading to dose modification are summarised in Table B.2-41.

Table B.2-41. AURORA 1: Most common TEAEs leading to dose modification (in ≥2% of patients in any group)

Abbreviations: TEAE = treatment-emergent adverse event Source: Otsuka 2020[109]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 90 of 165

==> picture [50 x 9] intentionally omitted <==

Adverse events of special interest

Known adverse effects of the CNIs ciclosporin and tacrolimus include kidney dysfunction, hypertension, electrolyte disturbances, tremor, and diabetes. Therefore, these events were of particular interest in this study.[109]

Hypertension occurred at a higher incidence in the voclosporin arm (20.2% vs 8.4% for placebo).[109] Consistent with the protocol guidance to maintain normal blood pressure through the use of antihypertensives, more patients in the voclosporin arm than the placebo arm were prescribed calcium channel blockers (33% vs 21%) and beta blockers (18% vs 11%) during the study; a similar proportion of patients in each arm (32% and 30%, respectively) were treated with diuretics. The majority of hypertension events were mild or moderate. Overall, there was no significant difference in mean blood pressure between the treatment groups.

No voclosporin-treated patients recorded TEAEs of diabetes or hyperglycemia (vs one of each event in the placebo arm).[2] A total of 18 (10%) patients in each treatment group had a confirmed eGFR decrease (prespecified as a > 30% decrease from baseline) at any time throughout the study. Only 2% of patients in each treatment group discontinued study drug due to eGFR decrease, which suggests that the eGFR decreases were largely reversible in both treatment groups.[2] Incidence of investigator-reported serious renal dysfunction was low and similar between treatment groups (voclosporin, 3%; placebo, 2%). Mean systolic blood pressure increased by 3.9 mmHg in the voclosporin group at week 2 and returned to baseline levels by week 8.

==> picture [51 x 10] intentionally omitted <==

AURORA 1 safety conclusions

Voclosporin was well tolerated in the AURORA 1 study with no new or unexpected safety signals observed.[109] Three placebo patients died as a result of TEAEs. An additional two patients in the placebo group and one patient in the voclosporin group died due to AEs which started more than 30 days after the last dose of study drug. A similar proportion of patients in each arm experienced serious TEAEs (20.8% in the voclosporin arm and 21.3% in the placebo arm) or had their study treatment discontinued as a result of a TEAE (11.2% and 14.6%, respectively).

The safety profile of voclosporin was comparable with that of the placebo on a background of MMF and low-dose steroids in this 52-week trial. The AEs observed in both treatment groups were as expected for the population and treatment regimen.[2]

B.2.10.3 AURORA 2 Phase 3 long-term continuation study

==> picture [49 x 10] intentionally omitted <==

Adverse events

The primary objective of the AURORA 2 study was to evaluate the long-term safety and tolerability of continued treatment with voclosporin for up to three years.[113] Throughout the study, voclosporin was well tolerated with no new or unexpected safety signals observed. The overall profile of adverse events seen in the second and third years of treatment was similar to that seen in the first year (AURORA 1), although the frequency of AEs reduced each year. A summary of TEAEs reported during AURORA 2 is in Table B.2-42.[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

Table B.2-42. Summary of TEAEs reported in AURORA 2

Abbreviations: TEAE = treatment-emergent adverse event Source: AURORA 2 CSR[113]

==> picture [50 x 10] intentionally omitted <==

Commonly reported adverse events

The most commonly reported AEs in AURORA 2 were infections and was consistent with findings from the AURORA 1 study.[113] Infections were reported by xx% of patients in the voclosporin arm and xx% of patients in the placebo arm (see Table B.2-43). Given the study population was immunosuppressed, an expected wide variety of infections were reported; with most frequent infections in the voclosporin arm being UTIs, URTIs, and viral URTIs. Coronavirus infections and herpes zoster were more common in the placebo arm.[113] Most infections were of mild or moderate intensity, with only three patients in each study arm recording severe infections (viral URTI, coronavirus and breast abscess in the voclosporin arm; three events of coronavirus in the placebo arm).[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 92 of 165

Table B.2-43. Summary of TEAEs reported by ≥3% of patients in either arm (AURORA 2)

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 93 of 165

Abbreviations: TEAE = treatment-emergent adverse event Source: AURORA 2 CSR[113]

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis © Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved Page 94 of 165

==> picture [50 x 10] intentionally omitted <==

Serious adverse events

There were more SAEs in the placebo arm than the voclosporin arm during AURORA 2 (xx patients [xxxx%] versus xx patients [xxxx%]).[113] Infections were the most frequently reported SAE, with the predominant cause being coronavirus infections; reported by five patients in the placebo arm (xxx%) and two patients (xxx%) in the voclosporin arm (see Table B.2-44).[113]

The AURORA 2 investigators considered only three SAEs to be related to study treatment, namely disseminated tuberculosis and hypertension in placebo-treated patients and URTI in a voclosporin-treated patient.[113] More patients in the placebo arm than the voclosporin arm experienced SAEs that were considered to be related to their disease, most commonly worsening LN (xxx% versus xxx% respectively), SLE flare (xxx% versus xxx%) and osteonecrosis (xxx% versus x%). One patient in the voclosporin arm recorded an SAE of decreased GFR.[113]

Table B.2-44. Summary of serious TEAEs occurring in >1% of patients in either treatment arm (AURORA 2)

Abbreviations: TEAE = treatment-emergent adverse event Source: AURORA 2 CSR[113]

==> picture [50 x 10] intentionally omitted <==

Deaths

xxxx patients died during the study, all of whom were in the placebo arm and three of which were due to SARS-CoV-2 coronavirus infection.[113] xxxxxxxxxxxxxxxxxxx was due to a pulmonary embolism. xxxxx events were treatment-emergent and none of the events were considered by the study investigators to be related to study treatment. In the case of the pulmonary embolism, the investigator considered it to be related to LN disease.[113]

==> picture [51 x 9] intentionally omitted <==

Adverse events leading to treatment discontinuation

More patients in the placebo arm than the voclosporin arm (xxx[xxxx%] versus xxx[xxx%], respectively) had their study treatment discontinued permanently as a consequence of an AE (see Table B.2-45).[113] The most common AEs leading to treatment discontinuation were worsening LN (xxx% versus xxx%), decreased GFR (xxx% versus xxx%), SLE flare (xxx% versus xxx%) and renal impairment (xxx% versus xxx%) in the placebo and voclosporin arms,

Company evidence submission template for voclosporin with immunosuppressive therapies for treating lupus nephritis

© Otsuka Pharmaceuticals (U.K.) Ltd. (2022). All rights reserved

respectively. Infections causes xxxx patients in the placebo arm to stop treatment and xxx patient in the voclosporin arm.[113]

Table B.2-45. TEAEs leading to discontinuation of voclosporin or placebo