==> picture [215 x 22] intentionally omitted <==

Single Technology Appraisal

Daratumumab in combination for untreated systemic amyloid light-chain amyloidosis [ID3748]

Committee Papers

==> picture [215 x 22] intentionally omitted <==

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

SINGLE TECHNOLOGY APPRAISAL

Daratumumab in combination for untreated systemic amyloid light-chain amyloidosis [ID3748]

Contents:

The following documents are made available to consultees and commentators:

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

1. Company submission from Janssen

2. Clarification questions and company responses

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

• a. Myeloma UK b. UK Kidney Association

4. Evidence Review Group report prepared by York Centre for Reviews and Dissemination – factual accuracy check

5. Technical engagement response from company

6. Technical engagement responses and statements from experts: a. Dr Carol Whelan, clinical expert, nominated by British Society for Heart Failure

• b. Dr Charlotte Manisty, clinical expert, nominated by British Cardiovascular Society

• c. Dr Jennifer Pinney, clinical expert, nominated by UK Kidney Association

• d. Huw Stiley, patient expert, nominated by Myeloma UK

7. Technical engagement responses from consultees and commentators: a. Myeloma UK b. British Society for Heart Failure

8. Evidence Review Group critique of company response to technical engagement prepared by York Centre for Reviews and Dissemination

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

Daratumumab in combination for newly diagnosed systemic amyloid light-chain amyloidosis [ID3748]

Document B

Company evidence submission

June 2021

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Instructions for companies

This is the template for submission of evidence to the National Institute for Health and Care Excellence (NICE) as part of the single technology appraisal (STA) process. Please note that the information requirements for submissions are summarised in this template; full details of the requirements for pharmaceuticals and devices are in the user guide.

This submission must not be longer than 150 pages, excluding appendices and the pages covered by this template. If it is too long it will not be accepted.

Companies making evidence submissions to NICE should also refer to the NICE guide to the methods of technology appraisal and the NICE guide to the processes of technology appraisal.

In this template any information that should be provided in an appendix is listed in a box.

Highlighting in the template (excluding the contents list)

Square brackets and grey highlighting are used in this template to indicate text that should be replaced with your own text or deleted. These are set up as form fields, so to replace the prompt text in [grey highlighting] with your own text, click anywhere within the highlighted text and type. Your text will overwrite the highlighted section.

To delete grey highlighted text, click anywhere within the text and press DELETE.

Grey highlighted text in the footer does not work as an automatic form field, but serves the same purpose – as prompt text to show where you need to fill in relevant details. Replace the text highlighted in [grey] in the header and footer with appropriate text. (To change the header and footer, double click over the header or footer text. Double click back in the main body text when you have finished.)

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Contents

Contents ........................................................................................................................................ 3 List of tables .................................................................................................................................. 5 List of figures ................................................................................................................................. 7 List of abbreviations ...................................................................................................................... 9 Introduction ................................................................................................................................. 13 B.1 Decision problem, description of the technology and clinical care pathway .......................... 14 Decision problem ........................................................................................................... 15 Description of the technology being appraised .............................................................. 21 Health condition and position of the technology in the treatment pathway .................... 24 Disease overview .................................................................................................... 24 Impact of AL amyloidosis on patients ..................................................................... 27 Description of the clinical care pathway .................................................................. 29 Equality considerations .................................................................................................. 32 B.2 Clinical effectiveness ............................................................................................................ 34 Identification and selection of relevant studies ............................................................... 35 List of relevant clinical effectiveness evidence ............................................................... 35 Summary of methodology of the relevant clinical effectiveness evidence ..................... 37 Trial design ............................................................................................................. 37 Trial methodology ................................................................................................... 38 Baseline characteristics .......................................................................................... 43 Statistical analysis and definition of study groups in the relevant clinical effectiveness evidence .................................................................................................................................. 46 Trial populations ..................................................................................................... 46 Statistical methods .................................................................................................. 47 Participant flow in the relevant randomised controlled trials ................................... 49 Quality assessment of the relevant clinical effectiveness evidence ............................... 49 Clinical effectiveness results of the relevant trials .......................................................... 50 Haematologic response .......................................................................................... 53 Major organ deterioration progression-free survival (MOD-PFS) ............................ 58 Overall survival ....................................................................................................... 62 Time to subsequent non-cross resistant anti-plasma cell therapy .......................... 64 Cardiac, renal and liver responses ......................................................................... 69 Health-related quality of life .................................................................................... 74 Subgroup analysis ......................................................................................................... 76 Meta-analysis ................................................................................................................. 80 Indirect and mixed treatment comparisons .................................................................... 80 Adverse reactions ........................................................................................................ 81 Treatment duration and dosage ............................................................................ 81 Adverse events ..................................................................................................... 83 Ongoing studies ........................................................................................................... 91 Innovation .................................................................................................................... 91 Interpretation of clinical effectiveness and safety evidence ......................................... 93 Strengths and limitations of the evidence base ..................................................... 93 B.3 Cost effectiveness ................................................................................................................ 96 Published cost-effectiveness studies ............................................................................. 97 Economic analysis ......................................................................................................... 97 Patient population ................................................................................................... 98 Model structure ....................................................................................................... 98 Intervention technology and comparators ............................................................. 105 Clinical parameters and variables ................................................................................ 106 Baseline characteristics ........................................................................................ 106 Efficacy in the decision tree .................................................................................. 106 Efficacy in the Markov model ................................................................................ 107

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Safety ................................................................................................................... 123 Measurement and valuation of health effects .............................................................. 123 Health-related quality-of-life data from clinical trials .............................................. 123 Mapping ................................................................................................................ 125 Health-related quality-of-life studies ...................................................................... 125 Adverse reactions ................................................................................................. 125 Health-related quality-of-life data used in the cost-effectiveness analysis ............ 128 Cost and healthcare resource use identification, measurement and valuation ............ 128 Intervention and comparators’ costs and resource use ........................................ 129 Health-state unit costs and resource use .............................................................. 132 Adverse reaction unit costs and resource use ...................................................... 139 Miscellaneous unit costs and resource use .......................................................... 139 Summary of base-case analysis inputs and assumptions ........................................... 140 Summary of base-case analysis inputs ................................................................ 140 Assumptions ......................................................................................................... 143 Base-case results ........................................................................................................ 147 Base-case incremental cost-effectiveness analysis results .................................. 147 Sensitivity analyses ..................................................................................................... 148 Probabilistic sensitivity analysis ............................................................................ 148 Deterministic sensitivity analysis ........................................................................... 149 Scenario analysis .................................................................................................. 150 Summary of sensitivity analyses results ............................................................... 153 Subgroup analysis ....................................................................................................... 153 Validation ................................................................................................................... 153 Validation of cost-effectiveness analysis ............................................................ 153 Interpretation and conclusions of economic evidence ............................................... 155 B.4 References ......................................................................................................................... 158

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

List of tables

Table 1: The decision problem .................................................................................................... 16 Table 2: Technology being appraised ......................................................................................... 21 Table 3: Daratumumab SC dosing schedule for AL amyloidosis in combination with bortezomib, cyclophosphamide and dexamethasone (BCd), four-week cycle dosing regimen. ...................... 23 Table 4: Dosing regimens for cyclophosphamide, bortezomib and dexamethasone in the ANDROMEDA trial ...................................................................................................................... 23 Table 5: Prevalence of AL amyloidosis in the UK ........................................................................ 26 Table 6: Prognostic factors and thresholds for each stage of the Mayo Clinic Cardiac Staging System ........................................................................................................................................ 27 Table 7: Clinical effectiveness evidence from ANDROMEDA ..................................................... 36 Table 8: Summary of the ANDROMEDA trial methodology ......................................................... 38 Table 9: Summary of haematologic response endpoints and definitions based on the consensus guidelines .................................................................................................................................... 42 Table 10: Summary of organ response and progression criteria based on the consensus guidelines .................................................................................................................................... 43 Table 11: Baseline patient characteristics in ANDROMEDA (ITT population) ............................. 43 Table 12: Baseline patient disease characteristics in ANDROMEDA (ITT population) ............... 45 Table 13: Trial populations used for the analysis of endpoints of ANDROMEDA ........................ 46 Table 14: Statistical methods for the primary analysis of ANDROMEDA .................................... 47 Table 15: Quality assessment of the ANDROMEDA trial ............................................................ 49 Table 16: Summary of ANDROMEDA data cuts ......................................................................... 51 Table 17: Summary of overall best confirmed haematologic response based on IRC assessment; ITT analysis set (14[th ] February 2020 data cut-off and 13[th] November 2020 data cut-off) ............ 54 Table 18: Summary of confirmed CHR at six- and 12-months based on IRC assessment, ITT analysis set (14[th] February 2020 data cut-off and 13[th] November 2020 data cut-off) .................. 56 Table 19: Summary of time to haematologic response based on IRC assessment; haematologic response-evaluable analysis set (14[th] February 2020 data cut-off and 13[th] November 2020 data cut-off) ......................................................................................................................................... 57 Table 20: Summary of duration of CHR based on IRC assessment; responders in ITT analysis set (14[th] February 2020 data cut-off) ........................................................................................... 58 Table 21: Summary of primary, sensitivity and supplementary analysis of MOD-PFS based on IRC assessment; ITT analysis set (14[th] February 2020 data cut-off) .......................................... 60 Table 22: Summary of OS; ITT analysis set (14[th] February 2020 data cut-off) ........................... 63 Table 23: Summary of subsequent non-cross resistant, anti-plasma cell therapy by therapeutic class, pharmacologic class, and preferred term; safety analysis set (14[th] February 2020 data cutoff) ............................................................................................................................................... 65 Table 24: Summary of time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (14[th] February 2020 and 13[th] November 2020 data cut-off) ..................................... 67 Table 25: Cardiac and renal six-month response rates (14[th] February 2020 data cut-off) .......... 71 Table 26: Summary of cardiac response rate at 6, 12 and 18 months based on IRC assessment without censoring non-cross resistant anti-plasma cell therapy; cardiac response-evaluable analysis set (13[th] November 2020 data cut-off) ........................................................................... 72 Table 27: Summary of renal response rate at 6, 12 and 18 months based on IRC assessment without censoring non-cross resistant anti-plasma cell therapy; renal response-evaluable analysis set (13[th] November 2020 data cut-off) ........................................................................... 72 Table 28: Median time to cardiac, renal and liver response based on IRC assessment (14[th] February 2020 data cut-off) ......................................................................................................... 73 Table 29: Cardiac, renal and liver progression rates at six months based on IRC assessment (14[th] February 2020 data cut-off) ................................................................................................. 74 Table 30: Summary of exposure to study treatment, safety analysis set (14[th] February 2020 data cut-off) ......................................................................................................................................... 83 Table 31: Overall summary of treatment-emergent adverse events; safety analysis set (14[th] February 2020 data cut-off) ......................................................................................................... 85

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Table 32: Most commonly reported (>10% in either arm) treatment-emergent adverse events by preferred term; safety analysis set (14[th] February 2020 data cut-off) .......................................... 86 Table 33: Most commonly reported (>5%) toxicity Grade 3 or 4 treatment-emergent adverse events by system organ class and preferred term; safety analysis set (14[th] February 2020 data cut-off) ......................................................................................................................................... 87 Table 34: Most common (at least 2% in either arm) treatment-emergent serious adverse events by system organ class and preferred term; safety analysis set (14[th] February 2020 data cut-off)87 Table 35: Number of patients with treatment-emergent infusion-related reactions by system organ class, preferred term and maximum toxicity grade; safety analysis set (14[th] February 2020 data cut-off) ................................................................................................................................. 88 Table 36: Most commonly reported (≥5% in either arm) toxicity Grade 3 or 4 treatment-emergent adverse events by system organ class and preferred term; safety analysis set (13[th] November 2020 data cut-off) ........................................................................................................................ 90 Table 37: Comparison of patient characteristics between ANDROMEDA and EMN23 ............... 94 Table 38: Features of the economic analysis ............................................................................ 102 Table 39: Baseline characteristics for the base case population ............................................... 106 Table 40: Haematologic response distribution over six months within the decision tree (base case analysis) .................................................................................................................................... 106 Table 41: Haematologic response distribution over three months within the decision tree ....... 107 Table 42: Model fit statistic for OS curve extrapolations for patients with NR or PR from Palladini et al., (2012) (six-month landmark) ........................................................................................... 110 Table 43: Model fit statistic for OS curve extrapolations for patients with VGPR from Palladini et al., (2012) (six-month landmark) ............................................................................................... 112 Table 44: Model fit statistic for OS curve extrapolations for patients with CR from Palladini et al., (2012) (six-month landmark) ..................................................................................................... 113 Table 45: Mortality distribution by health state for cycles four to six (ANDROMEDA; 14[th] February 2020 data cut-off) ...................................................................................................................... 115 Table 46: Mortality distribution by health state for cycles seven and beyond (ANDROMEDA; 14[th] February 2020 data cut-off) ....................................................................................................... 115 Table 47: Values informing transition probabilities to ‘End-stage organ failure’ by haematologic response ................................................................................................................................... 118 Table 48: Transition probabilities stratified by haematologic response for cycles 3–6 .............. 120 Table 49: Transition probabilities stratified by haematologic response for cycles 7 and beyond121 Table 50: Number of treatment cycles received in the base case and scenario analyses ........ 122 Table 51: Adverse event probabilities used in the model base case ......................................... 123 Table 52: Utility values for haematologic response derived from the ANDROMEDA trial ......... 123 Table 53: Utility values for haematologic response at baseline line and post-treatment derived from expert clinicians ................................................................................................................ 124 Table 54: Progression-related health state utility values ........................................................... 124 Table 55: End stage organ failure utility decrements ................................................................ 125 Table 56: Adverse event utility decrements and durations ........................................................ 127 Table 57: Total adverse event disutilities by treatment arm ...................................................... 127 Table 58: Summary of utility values for cost-effectiveness analysis .......................................... 128 Table 59: DBCd and BCd dosing regimens ............................................................................... 129 Table 60: Mean relative dose intensities ................................................................................... 129 Table 61: First-line drug acquisition costs ................................................................................. 130 Table 62: Drug costs cycle ........................................................................................................ 130 Table 63: Drug administration routes ........................................................................................ 131 Table 64: Administration unit costs ........................................................................................... 131 Table 65: First-line co-medications per drug regimen ............................................................... 132 Table 66: First-line co-medication unit costs ............................................................................. 132 Table 67: Disease monitoring costs per unit ............................................................................. 133 Table 68: Monitoring frequency of resource use by Markov health state .................................. 134 Table 69: Healthcare resource use unit costs ........................................................................... 135 Table 70: Healthcare resource use frequencies in the Markov model ....................................... 136 Table 71: Second-line treatment regimen acquisition costs (base case) .................................. 137 Table 72: Second-line treatment regimen duration and cost (acquisition and administration) .. 137

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Table 73: Third-line treatment regimen acquisition costs (scenario analysis) ........................... 137 Table 74: Third-line treatment regimen duration and cost (acquisition and administration) ...... 138 Table 75: Recurring treatments for end-stage organ failure unit costs ...................................... 138 Table 76: One-off treatments for end-stage organ failure unit costs ......................................... 138 Table 77: Recurring treatments for end-stage organ failure resource use frequencies ............. 139 Table 78: One-off treatments for end-stage organ failure resource use frequencies ................ 139 Table 79: Adverse event unit costs ........................................................................................... 139 Table 80: End of life costs ......................................................................................................... 140 Table 81: Summary of variables applied in the economic model base case ............................. 140 Table 82: List of assumptions for the base case analysis ......................................................... 144 Table 83: Base case results ...................................................................................................... 147 Table 84: Probability of cost-effectiveness at a WTP threshold of £20,000 and £30,000 .......... 147 Table 85: Probabilistic base case results .................................................................................. 148 Table 86: Summary of scenario analyses ................................................................................. 150 Table 87: Scenario analyses results ......................................................................................... 152 Table 88: Predicted survival at 12 months by haematologic response (six-month landmark) ... 154 Table 89: Predicted survival at 24 months by haematologic response (six-month landmark) ... 154 Table 90: Predicted survival at 36 months by haematologic response (six-month landmark) ... 154

List of figures

Figure 1: Summary of the mechanism of action of daratumumab in systemic AL amyloidosis ... 22 Figure 2: Current and expected pathway of care for AL amyloidosis patients in UK clinical practice........................................................................................................................................ 31 Figure 3: Design of the ANDROMEDA study .............................................................................. 37 Figure 4: Overview of ANDROMEDA daratumumab dosing schedule ........................................ 38 Figure 5: Kaplan-Meier plot of MOD-PFS based on IRC assessment, IPCW analysis; ITT analysis set (14[th] February 2020 data cut-off) ............................................................................. 60 Figure 6: Weighted Kaplan-Meier plot of MOD-EFS based on IRC assessment; ITT analysis set (14[th] February 2020 data cut-off) ................................................................................................. 62 Figure 7: Kaplan-Meier plot for OS; ITT analysis set (14[th] February 2020 data cut-off) .............. 63 Figure 8: Kaplan-Meier plot for time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (14[th] February 2020 data cut-off) ......................................................... 68 Figure 9: Kaplan-Meier plot for time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (13[th] November 2020 data cut-off) ....................................................... 69 Figure 10: Mean EQ-5D-5L utility scores over time; ITT analysis set (14[th] February 2020 data cut-off) ......................................................................................................................................... 75 Figure 11: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 1 of 2) (14[th] February 2020 data cut-off).......................................................... 77 Figure 12: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 2 of 2) (14[th] February 2020 data cut-off).......................................................... 78 Figure 13: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 1 of 2) (13[th] November 2020 data cut-off) ....................................................... 79 Figure 14: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 2 of 2) (13[th] November 2020 data cut-off) ....................................................... 80 Figure 15: The multiple mechanisms of actions of daratumumab in AL amyloidosis .................. 92 Figure 16: Cost-effectiveness model structure ............................................................................ 99 Figure 17: OS curve extrapolations for patients with PR from Palladini et al., (2012) (six-month landmark) .................................................................................................................................. 109 Figure 18: OS curve extrapolations for patients with NR from Palladini et al., (2012) (six-month landmark) .................................................................................................................................. 110 Figure 19: Weighted PR and NR OS curve extrapolations from Palladini et al., (2012) (six-month landmark) .................................................................................................................................. 111 Figure 20: OS curve extrapolations for patients with VGPR from Palladini et al., (2012) (sixmonth landmark) ....................................................................................................................... 112

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Figure 21: OS curve extrapolations for patients with CR from Palladini et al., (2012) (six-month landmark) .................................................................................................................................. 113 Figure 22: OS curve extrapolations stratified by haematologic response from Palladini et al ., 2012 (six-month landmark) ........................................................................................................ 114 Figure 23: Time-to-MOD-PFS by hematologic response........................................................... 117 Figure 24: Extrapolated time-to-MOD-PFS curves .................................................................... 118 Figure 25: Time to subsequent non-cross resistant anti-plasma cell therapy stratified by hematologic response ............................................................................................................... 119 Figure 26: Extrapolated time-to-subsequent non-cross resistant anti-plasma cell therapy ....... 120 Figure 27: Probabilistic ICER convergence plot ........................................................................ 148 Figure 28: Cost effectiveness plane scatterplot ......................................................................... 149 Figure 29: Cost-effectiveness acceptability curve ..................................................................... 149 Figure 30: Tornado plot (ICER) ................................................................................................. 150

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

List of abbreviations

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Introduction

Patients with systemic amyloid light chain (AL) amyloidosis experience poor prognosis, with an estimated four-year survival rate of 54%, and almost a third of patients die within one year of diagnosis.[1, 2] In current UK clinical practice, patients receive light chain suppressive chemotherapy, however, the majority of patients fail to achieve a complete haematologic response (CHR), which is the main aim of treatment, and many patients experience relapse with progression of organ damage.[3-6] There is a clear unmet need for an effective, well-tolerated treatment option to support the management of patients with this condition.

Daratumumab in combination with cyclophosphamide, bortezomib and dexamethasone received European marketing authorisation for the treatment of adult patients with newly diagnosed AL amyloidosis on 21[st] June 2021.[7] Marketing authorisation for this indication with the MHRA is expected in *********** ****, following the reliance route.

This submission presents the clinical and cost-effectiveness evidence for the use of daratumumab SC in combination with cyclophosphamide, bortezomib and dexamethasone within its marketing authorisation for the treatment of newly diagnosed patients with AL amyloidosis. The pivotal clinical trial, ANDROMEDA, referred to the cyclophosphamide, bortezomib and dexamethasone regimen as “CyborD”, but “BCd” is used throughout this submission to align with the terminology used in the NICE final scope.[8]

Overall, the data presented herein show that daratumumab SC in combination with BCd is superior to BCd alone in achieving deep, rapid haematologic and organ responses for patients with AL amyloidosis. In the ANDROMEDA trial, the proportion of patients achieving a CHR, and the rate at which they did so, was statistically significantly higher in the daratumumab SC with bortezomib, cyclophosphamide and dexamethasone (DBCd) treatment arm as compared with the BCd arm, and organ response rates were almost two-fold greater for patients treated with DBCd as compared with BCd alone. This was consistent across all pre-specified subgroups. The safety profile of DBCd was broadly consistent with the established safety profiles of daratumumab and BCd: TEAEs were generally manageable with no new safety concerns identified which is particularly important for patients when increasing the number of regimens within a combination therapy. Results from a de novo cost-utility analysis indicate that the introduction of DBCd to UK clinical practice represents a cost-effective use of NHS resources, with an incremental cost-effectiveness ratio (ICER) versus BCd alone falling below the willingness-to-pay (WTP) threshold of £30,000. If recommended, DBCd would be the first treatment specifically licensed for AL amyloidosis patients in the UK, representing a greatly needed step-change in the care available for these patients.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

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

Summary of decision problem, the technology and clinical care pathway Disease overview

• AL amyloidosis is a rare and debilitating condition and is the most frequent and severe form of amyloidosis in which amyloid fibrils are deposited in organs around the body. This causes debilitating symptoms which affect patients’ quality of life, leading to significant morbidity, organ dysfunction and death.

• An annual incidence 1 in 100,000 people in the UK annually has been estimated, with incidence known to increase with age and to be higher amongst males.[9-12] AL amyloidosis typically presents systemically (93% of AL amyloidosis patients).

• The symptoms of AL amyloidosis are often non-specific, such as weight loss and fatigue, contributing to prolonged times for diagnosis and frequent misdiagnosis, with more severe symptoms developing as the condition progresses.

• The heart and kidneys are the most commonly affected organs and the type, number and extent of organ involvement is a key factor in influencing survival of these patients.[1, 13, 14] The Mayo Clinic Staging System is the most widely used staging system for AL amyloidosis, with prognosis worsening as patients progress through more advanced stages.[14-17]

•  Over time, almost all patients experience progression of organ involvement and ultimately death, with heart failure representing the most common cause of mortality in these patients.[18-20]

• Impact on patients

• Patients with AL amyloidosis suffer both physical and psychological burdens: from the wideranging symptoms of AL amyloidosis and the side effects associated with off-label chemotherapies, to the anxiety, depression and low self-worth associated with diagnosis of a life-limiting disease with a poor prognosis and no licensed treatment options; AL amyloidosis patients experience a significantly reduced quality of life.

• These burdens are exacerbated by prolonged wait times for a correct diagnosis, due in part to the rarity of the disease and lack of clinical knowledge surrounding it. This delayed diagnosis is associated with further disease progression and corresponding worsened prognoses.

• Patients who reach end-stage kidney failure become reliant on regular dialysis, which has a further substantial impact on patient quality of life.[21-23]

• Current clinical care

• As AL amyloidosis is incurable, the primary therapeutic goal of its treatment is to achieve a rapid, deep and durable haematologic response.[24, 25]

• Before daratumumab, there were no therapies specifically licensed for the treatment of patients with AL amyloidosis; furthermore, there are no NICE guidelines currently available for its treatment.

• The majority (*****%) of newly diagnosed patients are treated with BCd as a first-line therapy in the UK.[88] As such, BCd represents the standard of care for these patients and constitutes the sole comparator in this appraisal.

Unmet need

• The majority of patients currently fail to achieve a complete haematologic response (CHR) through treatment with the current standard of care in the UK, and organ response rates are poor.[27-31]

• There remains a high level of unmet need for a novel, effective and well-tolerated treatment for newly diagnosed AL amyloidosis patients that has the potential to induce a rapid, deep and durable haematologic response. Introduction of such a treatment option would improve patient prognosis and health-related quality of life (HRQoL), delay organ failure and prolong survival.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Daratumumab SC

• Daratumumab is a fully human monoclonal antibody that binds to CD38 on the surface of haematologic cells, including the clonal plasma cells that produce amyloidogenic immunoglobulin light-chain, to reduce native light-chain production and the associated organ toxicity in patients with systemic AL amyloidosis.

• Daratumumab has been shown to be efficacious and safe in multiple myeloma (MM) and has a licence for use in four different treatment regimens in MM indications in the European Union.[32]

• The efficacy and safety of daratumumab SC in combination with BCd has been assessed in the ANDROMEDA trial. DBCd was superior to BCd in achieving deep, rapid haematologic and organ responses and was associated with an improvement in HRQoL over time and a tolerable safety profile that raised no new safety concerns.

==> picture [32 x 12] intentionally omitted <==

Decision problem

The objective of this appraisal is to determine the clinical and cost-effectiveness of DBCd, in line with its marketing authorisation, i.e. for the treatment of adult patients with newly diagnosed systemic light-chain (AL) amyloidosis.[32] The decision problem addressed in this submission, compared to that defined in the final scope issued by NICE, is summarised in Table 1.[8]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 1: The decision problem

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

==> picture [113 x 112] intentionally omitted <==

from the licensed indication for DBCd.[32]  These patients have high risk systemic AL amyloidosis and an extremely poor prognosis.[26]  It is Janssen’s view that it is important that any recommendation for DBCd in AL amyloidosis is not restricted in such a way to exclude patients with Stage IIIb disease, a group of more severe patients, who have an extremely poor prognosis and life expectancy.

Abbreviations: AE: adverse event; AL: amyloid light-chain; ASCT: autologous stem cell transplant; BCd: bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; CTd: cyclophosphamide, thalidomide and dexamethasone; DBCd: daratumumab SC, bortezomib, cyclophosphamide and dexamethasone; EMA: Europeans Medicines Agency; FDA: Food and Drugs Administration; HRQoL: health-related quality of life; Md: melphalan and dexamethasone; MOD-EFS: major organ deterioration event-free survival; MOD-PFS: major organ deterioration progression-free survival; NHS: National Health Service; NICE: National Institute for Health and Care Excellence; OS: overall survival; PAS: patient access scheme; PFS: progression-free survival; PSS: Personal Social Services QALY: quality-adjusted life year; UK: United Kingdom.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

==> picture [34 x 12] intentionally omitted <==

Description of the technology being appraised

A description of the technology being appraised, daratumumab SC in combination with BCd, is presented in Table 2.

Table 2: Technology being appraised

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

==> picture [102 x 269] intentionally omitted <==

Figure 1: Summary of the mechanism of action of daratumumab in systemic AL amyloidosis

==> picture [340 x 90] intentionally omitted <==

==> picture [340 x 90] intentionally omitted <==

Abbreviations: ADCC: antibody-dependent cell-mediated cytotoxicity; ADPC: antibody-dependent cellular phagocytosis; AL: amyloid light chain; CDC: complement-dependent cytotoxicity; NK: natural killer. Source: Janssen (Data on File): Daratumumab AL Amyloidosis Scientific Communications Platform.[40]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

o Sorbitol (E420) o Water for injections Method of AL amyloidosis posology administration Daratumumab 1,800 mg (15 mL vial; 120 mg daratumumab per mL) is and dosage available as a solution for subcutaneous (SC) injection administered over approximately 3–5 minutes according to the dosing schedule shown in Table 3.[32] As shown, in the ANDROMEDA trial from Week 25 onwards, daratumumab SC was administered every four weeks until disease progression or a maximum of 24 cycles (~2 years) from the first dose of treatment.[32, 35]

Table 3: Daratumumab SC dosing schedule for AL amyloidosis in combination with bortezomib, cyclophosphamide and dexamethasone (BCd), four-week cycle dosing regimen.

a In the clinical trial, daratumumab was given until disease progression or a maximum of 24 cycles (~2 years) from the first dose of study treatment. Source: Daratumumab SmPC.[32]

Daratumumab SC formulation is not intended for intravenous (IV) administration and should be given by subcutaneous injection only, using the doses specified.[32]

Drug administration should be performed by a healthcare professional, and the first dose should be administered in an environment where resuscitation facilities are available.[32]

It is anticipated that the concomitant medications of the regimen (cyclophosphamide, bortezomib and dexamethasone) will be administered to patients in line with the dosing schedules in the ANDROMEDA trial presented in Table 4. For further details of the additional components of the DBCd regimen, please refer to the respective SmPCs.[41-43] For further information regarding concomitant medications recommended to be administered alongside DBCd to manage the risk of infusion-related reactions, please see the daratumumab SmPC.[32]

Table 4: Dosing regimens for cyclophosphamide, bortezomib and dexamethasone in the ANDROMEDA trial

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations : ADCC: antibody-dependent cell mediated cytotoxicity; ADCDP: antibody dependent cellular phagocytosis; AL: amyloid light-chain; BCd: bortezomib, cyclophosphamide and dexamethasone; CDC: complement dependent cytotoxicity; DBd: daratumumab, bortezomib and dexamethasone; DBCd: daratumumab SC, bortezomib, cyclophosphamide and dexamethasone; DBMP: daratumumab, bortezomib, melphalan and prednisone; DLd: daratumumab, lenalidomide and dexamethasone; IgG1ĸ: immunoglobulin G1 kappa; IV: intravenous; mAb: monoclonal antibody; MDSC: myeloid-derived suppressor cell; MM: multiple myeloma; SC: subcutaneous; Tregs: regulatory T cells.

==> picture [34 x 12] intentionally omitted <==

Health condition and position of the technology in the

treatment pathway

==> picture [38 x 10] intentionally omitted <==

Disease overview

Background to amyloidosis

Amyloidosis comprises a group of rare plasma cell disorders characterised by the formation of amyloid fibrils due to protein misfolding. The particular protein that misfolds is specific to the type of amyloidosis, but a common characteristic of all amyloidosis subtypes is the extracellular aggregation and accumulation of these fibrils within organs, resulting in impaired organ function and premature mortality.[44-46]

Types of amyloidosis may be differentiated based on clinical characteristics such as the pattern of tissue involvement (systemic versus localised), the role of underlying disease (primary versus secondary) and heritability of the condition (acquired versus hereditary), or according to the target tissue involved, such as in cardiac amyloidosis in which the heart is affected.[46-49] However, current guidelines from the International Society of Amyloidosis (ISA) recommend that amyloidosis should primarily be classified according to the amyloid precursor involved, given that this classification more closely reflects the underlying biology of the disease than the clinical classifications.[48]

AL amyloidosis

The most common and severe form of amyloidosis is immunoglobulin (Ig) light-chain (AL) amyloidosis, a rare and debilitating condition caused by an abnormality in certain cells found in the bone marrow, called plasma cells. While plasma cells in healthy people produce normal proteins (called ‘light chain proteins’) to help protect the body from infection, patients with AL amyloidosis produce erroneous forms of these proteins which create amyloid deposits when they enter the bloodstream. These proteins aggregate into thread-like strings (amyloid fibrils) that cannot be cleared easily. Over time, amyloid fibrils build up as AL amyloid deposits in tissues and organs. This gradually stops the organs functioning properly, causing debilitating symptoms

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

and ultimately leading to death. Unlike some other types of amyloidosis, AL amyloidosis is not hereditary.[44-46, 50]

AL amyloidosis has an annual incidence of approximately one case in every 100,000 people in the UK and accounts for approximately 60% of all amyloidosis cases.[9] AL amyloidosis may present locally, but the vast majority (93%) of patients have systemic involvement in which several organs are involved. Throughout this submission, the term ‘AL amyloidosis’ is used in reference to systemic AL amyloidosis.

Immunoglobulin heavy chain translocation (11;14) is one of the most prevalent chromosomal abnormalities in patients with AL amyloidosis, having an influence on prognosis via modification of the response to treatment.[14, 51] Translocation t(11;14), which involves the immunoglobulin heavy chain and genes encoding cyclin D1, is the most prevalent, found in approximately 39– 57% of patients.[45, 52, 53] This translocation promotes proliferation of plasma cells and has been associated with a poor response to standard treatment.[54, 55]

The clinical presentation of amyloidosis varies according to the type, number and extent of organ involvement.[18, 19] In AL amyloidosis, the heart and kidneys are the most commonly affected organs, with approximately 50–70% of patients experiencing cardiac involvement, and up to 70% experiencing renal involvement.[18, 19, 56] Other sites that may be affected include the liver, gastrointestinal tract, soft tissue and peripheral nervous system, with most patients experiencing involvement across multiple organs.[1, 18, 57] Accordingly, patients often present with non-specific symptoms such as weight loss, fatigue, weakness, loss of appetite, bruising of ankles and legs, shortness of breath with minimal exertion, numbness, tingling or pain in hands or feet, blood pressure change, dizziness, GI symptoms such as diarrhoea or constipation, pain and/or kidney issues. This non-specificity of symptoms poses a challenge for diagnosis and can result in delays of several months or longer for an initial diagnosis.[58, 59] Although some symptoms, such as periorbital purpura and tongue enlargement, are specific to AL amyloidosis, they are less common, occurring in around 15% of patients.[59]

As the condition progresses, more severe symptoms develop, which may include heart failure. In addition, patients with kidney involvement may experience malabsorption, albuminuria and nephrotic-range proteinuria which impact the quality of life; if diagnosed late, kidney involvement can lead to end-stage renal failure.[18, 19, 57, 60]

The majority of patients with AL amyloidosis fail to achieve a CHR following standard therapy, and eventually, almost all patients experience haematologic relapse and progression of organ involvement, and ultimately death.[61]

Epidemiology

Systemic AL amyloidosis is a rare disease for which there is limited evidence on prevalence.[11, 12, ] 62-68 The main source of epidemiological information in the UK is the National Amyloidosis Centre (NAC). An analysis of patients in the UK NAC Database estimated a prevalence of 11,006 amyloidosis cases between 1987–2019, with AL amyloidosis cases representing 55% of the total (6,008). Another analysis of patients in England in the same database reported 174 cases of AL amyloidosis per 1 million individuals in 2008 and indicated that AL amyloidosis referral rates doubled during 2000 to 2008.[63, 66] These studies are summarised in Table 5.

Overall, the studies have generally reported low incidence rates of less than 1,000 new diagnoses each year. Acknowledging the rarity and severity of the disease and the absence of

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

authorised medicines for the treatment of this condition, daratumumab was granted orphan designation by the EMA in March 2020 for the treatment of AL amyloidosis.[69]

In general, the incidence of AL amyloidosis increases with age, with the highest rates reported in elderly patients (i.e. age ≥65 years).[10-12] Incidence is also higher among males, accounting for 54–58.5% of cases.[10, 11, 64]

Table 5: Prevalence of AL amyloidosis in the UK

a Total population (N) not reported for either study. b Systemic AL amyloidosis was reported as: “the most prevalent” type of systemic amyloidosis, although the proportion of patients with AL amyloidosis specifically was not reported. Although the NAC serves the entire population of England, the incidence of new referrals to the NAC decreases as distance from the NAC increases (R[2] = 0.64, P = 0.005). Abbreviations: AL: light chain; NAC: National Amyloidosis Centre.

Disease prognosis and staging

As previously indicated, prognosis is poor for patients with AL amyloidosis, with nearly 30% of patients dying within the first year of diagnosis and an estimated 4-year survival rate of 54%.[1, 2] A key factor influencing survival is the type and number of organs involved, and prognosis is particularly poor in patients with multiple organ involvement.[1, 13, 14] Five-year survival has been shown to be 91% for patients with isolated renal involvement, compared to 42% among patients with multiple-organ involvement (p<0.001).[18] Notably, heart failure is the leading cause of death in patients with AL amyloidosis, and the presence and extent of cardiac involvement is among the strongest indicators of mortality risk.[20] If untreated, patients with cardiac involvement have a median survival of just six months from onset of heart failure.[70]

The most widely used staging system for AL amyloidosis was developed by the Mayo Clinic group in 2004 and revised in 2012.[14-16] The Mayo Clinic Staging System stratifies patients according to soluble serum cardiac biomarkers, as well as other important prognostic factors.[14, 15] Levels of the cardiac biomarkers N-terminal pro-brain natriuretic peptide (NT-proBNP) and cardiac troponin T (TnT) form the basis of the staging system, with patients assigned to Stages I, II and III according to the number of these prognostic factors found to be elevated above defined thresholds.[14] In a 2013 revision to the Mayo system, Stage III patients were further categorised into Stages IIIa or IIIb based on whether the level of NT-proBNP was below (IIIa) or above (IIIb) 8,500 ng/L (Table 6).[16]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Table 6: Prognostic factors and thresholds for each stage of the Mayo Clinic Cardiac Staging System

Abbreviations: cTnT: cardiac troponin T; NT-proBNP: N-terminal prohormone brain natriuretic peptide. Source: Dispenzieri et al., (2004);[15] Wechalekar et al., (2013).[16]

Prognosis worsens as patients progress through more advanced Mayo stages; this is supported by data from the European Myeloma Network (EMN), which analysed the real-world treatment outcomes for 3,000 patients in 10 European countries, of which 38% were UK patients, during the period 2011 to 2018. In alignment with previous studies, the EMN23 study identified an inverse relationship between median survival and cardiac stage, as defined by the Mayo 2004 / European modification staging system: whilst median survival was not reached by patients with Stage I disease, patients with Stage II, IIIa and IIIb disease had a median survival of 67.0 months, 31.1 months and 4.5 months, respectively.[17 ]

Delays in diagnosis are also associated with the poor prognosis of patients with AL amyloidosis since patients are often diagnosed after the disease has progressed to more advanced stages.[38, ] 71 The EMN retrospective analysis found that many patients had advanced disease at the point of diagnosis, with 35% having Stage II disease at diagnosis, 22% with Stage IIIa and 16% with Stage IIIb.[72] Furthermore, a recent US claims analysis of 1,403 patients has shown that symptoms related to advanced disease progression appear <1 year before diagnosis, whilst the median time from onset of symptoms to diagnosis overall is 2.7 years.[73] Ultimately, delayed diagnosis in patients with AL amyloidosis significantly and adversely impacts survival rates because disease is more advanced with later diagnosis and irreversible organ damage has already occured.[59, 74]

==> picture [39 x 10] intentionally omitted <==

Impact of AL amyloidosis on patients

AL amyloidosis has a significant impact on the lives of patients afflicted with this disease. Patients suffer from substantially reduced quality of life due to physical burdens from the wideranging symptoms and complications associated with the condition, in addition to side effects from currently available chemotherapy regimens. Patients suffer further psychologically, with many patients reporting anxiety, depression and low self-worth. These factors are compounded by delayed diagnosis and a lack of information around the disease.

Given the rarity of AL amyloidosis, Janssen conducted a patient workshop consisting of two online focus groups on 7[th] and 14[th] April 2021 to understand the psychological and emotional impact of AL amyloidosis on patients and carers in the UK. The methods for the workshop, which gathered insights from six patients with AL amyloidosis and one carer of a patient, are presented in Appendix N, and the full report is available in the reference pack.[75]

The burden of AL amyloidosis from a physical and emotional perspective, as well as the impact of barriers to diagnosis and treatment, is discussed further below.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Physical and emotional burden experienced by AL amyloidosis patients

AL amyloidosis, and its associated treatments, impose a substantial burden on patients and their carers, with the disease and its symptoms taking a major toll on quality of life. For the large proportion of AL amyloidosis patients with cardiac involvement, impaired heart functioning leads to shortness of breath, fatigue, and eventual development of congestive heart failure. Such physical impairment significantly affects patients’ ability to conduct day-to-day activities and may lead to rapid onset of exhaustion. One participant at the patient workshop described his ability to walk as greatly impaired and his breathing as ‘appalling’; he described that he is unable to bend down and expressed that he gets exhausted quickly.[75] Beyond cardiac problems, the systemic nature of AL amyloidosis means that the symptom burden is high, with fatigue, weakness, weight loss, muscle atrophy and neuropathy reported as the other symptoms that have the greatest impact on daily life.[76]

The physical burden of AL amyloidosis is supported quantitatively by the US prospective AL Amyloidosis Patient HRQoL Study, which compared the HRQoL of 341 AL amyloidosis patients with that of matched general population controls. This study identified that patients with newly diagnosed AL amyloidosis reported significantly lower scores across all subscales on the Short Form 36 Health Survey Questionnaire (SF-36v2) (p<0.05 for all comparisons) compared to the general population.[77] The study also underscored the relationship between cardiac involvement and reduced HRQoL, with 178 patients with cardiac involvement demonstrating significantly lower SF-36v2 scores compared to the general population.[77] Further, the subgroup of patients who did not meet cardiac response targets (≥30 reduction in NT-proBNP levels) had significantly worse mean SF-36v2 scores across all subscales compared to patients who did meet this target.[78 ]

With regards to treatment, many patients at the workshop indicated that it was difficult to separate feelings of ill health because of the disease versus feelings of ill health from treatment.[75] As discussed below, there are currently no approved therapies for AL amyloidosis in the UK, and off-label chemotherapies used in current clinical practice are associated with adverse events.[26]

Beyond their physical health, patients with AL amyloidosis experience significant mental health challenges. Findings from the patient workshop show that living with AL amyloidosis has a negative impact on the mental health of patients, which has worsened in the last year by the ongoing COVID-19 pandemic, causing delays in treatment and diagnosis. Patients expressed feelings of low self-worth, frustration at their declining physical ability and distress at their loss of independence. One individual indicated that he struggled at the time of diagnosis, noting that he did not want to talk with family and friends about his disease for months and that “the transplant cancellation was the lowest point in my life.” The patients at the workshop were in agreement that the optimal treatment option would allow them to live as long as possible for as well as possible, without having to trade-off between the two.[75]

The impact of AL amyloidosis on patients’ mental health is further supported by a large US study of 1,226 patients. High proportions of patients who had completed the SF-36v2 reported experiencing symptoms of anxiety (47%) and depression (37%), with many reporting that this disrupted their ability to work and reduced the amount of time spent on work and other activities.[79] In a further qualitative study of 199 AL amyloidosis patients, 63.0% reported that their diagnosis had made them feel frightened, 31.0% reported feeling depressed and 25.5% felt hopeless.[76] This reflects the poor prognosis and expected survival of patients diagnosed with AL amyloidosis, as described previously.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Barriers to diagnosis and treatment

Patients with AL amyloidosis experience a difficult journey due to delayed diagnoses and the lack of available support and disease information.[76, 80] A survey conducted by the US Amyloidosis Research Consortium (ARC), completed by 533 patients, family members or carers, including respondents from the UK, reported that just 37% of patients had received a diagnosis within six months, with 20% and 10% of patients waiting more than two years and three years, respectively.[76]

Non-specific symptoms and healthcare system barriers have both been cited as sources of the difficulties in establishing a diagnosis of AL amyloidosis.[76] AL amyloidosis may go undiagnosed due to patient interpretation of non-specific symptoms, such as fatigue, shortness of breath and muscle aches, as low priority concerns, or attributable to other less serious health conditions or to advancing age.[76, 81] Healthcare barriers mean that patients can struggle to get their physicians to investigate their observed symptoms and that these are subsequently frequently misdiagnosed. In the ARC study, cardiologists were the most frequently seen physician after the primary physician/general practitioner (24.5%), despite the fact that they were typically unlikely to diagnose the condition (doing so in only 22.6% of cases), with correct diagnoses more likely to come from haematologists or oncologists (34.1%).[76] Nearly half of AL amyloidosis patients in the ARC study reported having received a misdiagnosis.[76]

Aligned with this, several UK-based patients at the workshop described their journey to diagnosis as challenging due to the generic nature of the symptoms, with one patient recalling having been seen by several practitioners, including a nephrologist, who diagnosed multiple myeloma, before receiving the correct diagnosis from a haematologist. Some patients experienced a time to diagnosis of up to two years.[75]

The rarity of AL amyloidosis and the lack of any currently licensed treatment options for patients in the UK may contribute to the extended waiting time for diagnosis faced by patients. Introduction of DBCd to clinical practice following recommendation by NICE is likely to raise awareness of this rare disease, thus improving diagnosis rates and positively impacting patient outcomes and survival rates as a result.

Summary

Overall, the physical, psychological and social burdens of AL amyloidosis contribute to significantly reduced quality of life in patients, which is exacerbated further by healthcare barriers such as prolonged diagnosis time and misdiagnosis.

==> picture [39 x 11] intentionally omitted <==

Description of the clinical care pathway

Treatment pathway for AL amyloidosis in the UK

As AL amyloidosis is incurable, the primary goal of treatment is to achieve a rapid, deep and durable haematologic response as this is associated with improved quality of life and prolonged survival.[24, 25] Haematologic response represents a key goal in medical society guidelines, which recommend at least VGPR as the target and preferably CHR.[3-6] An early and deep haematologic response has been established as a key prognostic factor for survival as demonstrated by studies assessing the impact of timing and depth of haematologic response in patients with newly diagnosed AL amyloidosis.[2, 82]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Currently, there are no therapies in the UK that are specifically licensed for the treatment of patients with AL amyloidosis. There are also no NICE guidelines currently available for treatment of AL amyloidosis and the latest UK guidelines from the British Society for Haematology (BSH) were published in 2014.[83] During a UK advisory board in April 2021, clinical experts agreed that the BSH guidelines may be used to inform the treatment of patients in UK clinical practice, but that advice from the NAC represents the most valuable resource to inform treatment in the UK at this time.[26]

The small, select group of AL amyloidosis patients who are eligible to undergo ASCT have greatly improved prognosis and survival outcomes.[15, 84, 85] ASCT can be a highly effective treatment option, but the proportion of patients who are eligible to receive this treatment option is small, with many patients ineligible due to their high comorbidity burden: eligibility criteria for ASCT vary by country, though patients are typically considered ineligible for ASCT if they have ≥2 affected organs, severe cardiac dysfunction, end-stage renal disease or high overall comorbidity burden.[6] Indeed, a treatment rate of approximately 5% can be estimated from data from the British Society of Blood and Marrow Transplantation and Cellular Therapy; this is likely to have been even lower in the year 2020/2021 due to an increased risk of infection associated with transplantation during the COVID-19 pandemic.[86, 87]

This low rate of treatment with ASCT is supported by feedback from a UK advisory board, in which clinicians suggested that <1% of AL amyloidosis patients receive ASCT as a first-line treatment option (i.e. without prior induction therapy), reflecting the advanced stage of disease at presentation of many patients in the UK.[26] The experts noted that only 10 to 12 patients have received ASCT without previous induction therapy in the last 10 years.[26] This indicates that many of the patients undergoing ASCT are not newly diagnosed systemic AL amyloidosis patients.

Accordingly, bortezomib-based regimens are considered to represent the mainstay of treatment for AL amyloidosis. This is supported by the EMN23 study, which found that, between 2011 to 2018, 75.0% of AL amyloidosis patients received first-line bortezomib-based regimens. Furthermore, ASCT was used in 10% and 2% of those younger or older than 65, respectively, mostly for patients at an earlier cardiac stage (14%, 4%, 3%, and 1% for Stages I, II, IIIa and IIIb, respectively).[17]

Based on the same clinical expert opinion, the standard of care for newly diagnosed patients in clinical practice in the UK is BCd, and approximately *****% of patients are treated with BCd as first-line therapy.[88]

Clinical expert opinion suggests that, in the UK, melphalan in combination with dexamethasone is rarely used.[26] Lenalidomide with dexamethasone is an option only for patients with neuropathy, but it is very rarely used in newly diagnosed patients; typically, this treatment option is likely to be used only in patients who struggle with, or are contraindicated to, bortezomib.[26]

In the second line setting, clinical feedback was that patients may be re-treated with bortezomibbased regimens, particularly if they had responded well in the first-line setting. Carfilzomib, lenalidomide and dexamethasone (KRd) was indicated to not be used due to high levels of toxicity.

Based on clinical expert opinion elicited at the advisory board, the current treatment pathway for AL amyloidosis patients in the UK, and the expected treatment pathway should DBCd be recommended in this indication, is presented in Figure 2.[26] In this figure, the proportions of

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

patients who would receive each treatment following BCd or DBCd, as estimated by these clinicians, are also presented.

Figure 2: Current and expected pathway of care for AL amyloidosis patients in UK clinical practice

==> picture [453 x 79] intentionally omitted <==

==> picture [453 x 78] intentionally omitted <==

a Given the small proportion of newly-diagnosed AL amyloidosis patient who receive ASCT, this treatment is not considered as a comparator in this appraisal. Abbreviations: ASCT: autologous stem cell transplant; BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC, bortezomib, cyclophosphamide and dexamethasone.

Unmet need

There is a high level of unmet need for patients with AL amyloidosis, with the disease causing progressive, irreversible damage to multiple organs that leads to significant HRQoL impairment, substantial treatment costs and a very poor long-term prognosis. Nearly one-third (30%) of patients die within the first year of diagnosis, with an estimated 4-year survival rate of just 54%.[2, ] 74 This burden is particularly high in patients with advanced Mayo Clinic Stage IIIa/IIIb disease, who represent approximately 55% of all presenting UK AL amyloidosis patients: median overall survival in patients with Mayo Clinic Stage IIIb at baseline is approximately 2.5–3.5 months as compared with about 9.5–25.9 months in patients with Mayo Clinical Stage IIIa.[72] This reflects that heart failure is currently the leading cause of death in patients with systemic AL amyloidosis, with a median survival of just six months from the onset of heart failure.[18-20] Survival rates are also particularly low for the large proportion of patients with multiple organ involvement.[18]

Few robust clinical trials have been conducted in patients with AL amyloidosis to date. As a result, available clinical evidence is primarily based on retrospective studies, with only a small number of Phase II or III trials. Available studies of bortezomib-based regimens (e.g. BCd) typically show high ORRs, however the majority of patients currently fail to achieve the primary therapeutic goal of a CHR. This in turn allows toxic amyloids to continue to build up in the organs.[27-29] Organ response rates (OrRRs), which indicate the effect of treatment on organ function, have also been shown to be poor for bortezomib-based therapies: across several studies, most AL amyloidosis patients receiving first-line bortezomib-based regimes have been observed to fail to achieve cardiac (13 to 29%) and renal responses (19 to 29%), increasing overall disease burden and mortality risk.[27, 28, 30, 31]

Limited efficacy and adverse events are still concerns with current therapy that may result in further HRQoL impairment.[89] Treatments that have been shown to be more effective, such as ASCT or organ transplantation, are restricted to small groups of eligible patients; unfortunately, AL amyloidosis patients often have multi-organ involvement which makes them unsuitably fit to undergo ASCT.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Additionally, most patients experience a protracted delay between symptom onset and initial diagnosis, meaning many patients have advanced disease at the time of diagnosis and are given a poor prognosis at this point.[1, 38] Aside from the more severe and life-limiting symptoms associated with later stages of disease, this delayed diagnosis and the lack of any licensed treatments for AL amyloidosis contribute to the psychological burdens of anxiety, depression and low self-worth reported by patients. It is likely that the availability of a treatment option recommended specifically for the treatment of AL amyloidosis in the UK would have the dual benefits of raising clinician and patient awareness of the disease and reducing the stress and anxiety associated with diagnosis.

Beyond the physical and emotional aspects of life with AL amyloidosis for patients, the direct treatment costs for the healthcare system are substantial. In particular, significant costs are associated with patients who require multiple lines of therapy. For patients who reach end-stage renal failure, renal replacement therapy is necessary; for patients who do not receive, or are not eligible for, a transplant, treatment is with dialysis. In addition to being extremely costly for the healthcare system, the substantial impact of dialysis on patient HRQoL is well-documented.[21-23] As such, a reduction in the proportion of patients who require this treatment would be an important benefit of DBCd treatment in AL amyloidosis patients with kidney involvement.

Accordingly, there is a substantial unmet need for a novel, effective and well-tolerated treatment for newly diagnosed AL amyloidosis patients that has the potential to induce a rapid, deep and durable haematologic response. By doing so, such a treatment will improve the poor prognosis associated with this disease, improve patient HRQoL, delay organ failure and prolong survival.

Daratumumab SC in combination with BCd

Daratumumab in combination with BCd has received marketing authorisation for the treatment of adult patients with newly diagnosed systemic AL amyloidosis.[32] In this positioning, the sole relevant comparator to DBCd is BCd alone.

The efficacy and safety of DBCd has been compared directly to BCd in ANDROMEDA, a pivotal, randomised controlled trial of newly diagnosed AL amyloidosis patients. Within ANDROMEDA, DBCd has demonstrated a rapid and deep haematologic response, as well as high rates of cardiac and renal response, relative to BCd. The methodology and results of ANDROMEDA are presented in Section B.2.

NICE recommendation of DBCd as a treatment in this population in England and Wales would make it the first recommended treatment for AL amyloidosis patients specifically and would represent a step-change to patient care. The introduction of DBCd would fulfil a significant unmet need for a group of patients who suffer from a dearth of effective and tolerable treatment options and face an extremely limited prognosis and life expectancy.

==> picture [34 x 12] intentionally omitted <==

Equality considerations

In the ANDROMEDA study, patients with Stage IIIb cardiac disease were excluded during the screening period from participating in the trial, as they are not typically candidates for BCd at the specific dose and dosing schedule used in the trial.[35] It is important to note, however, that between the time of screening and of the first study drug administration, eight patients progressed to cardiac Stage IIIb (six in the BCd arm; two in the DBCd arm).

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

As described above, patients with Stage IIIb cardiac disease have the most severe degree of cardiac involvement.[16] However, clinical expert opinion supports that patients would be treated with DBCd in clinical practice should DBCd be recommended for use in the UK, as this would fulfil a significant unmet need in these patients. Furthermore, clinical experts emphasised that Stage IIIb patients comprise at least 20% of the AL amyloidosis patients observed in UK clinical practice and that they would expect such patients to derive clinical benefit should DBCd be recommended by NICE.[26] Patients with Stage IIIb disease are not excluded from the licensed indication for DBCd.[32]

These patients have high risk systemic AL amyloidosis and an extremely poor prognosis: in the EMN23 study, patients with Stage IIIb cardiac disease had a median survival of just 4.5 months.[17] These patients with severe cardiac involvement were further identified to receive bortezomib-based therapies often, with 81% and 8% of Stage IIIa and IIIb patients reported to receive these therapies, respectively.[17] Results of an analysis for patients with cardiac Stage IIIb disease which aimed to evaluate the safety profile of daratumumab monotherapy in this high-risk population found that from 14 patients who had received the first dose of daratumumab at least three months prior to the cut-off date, 9 (64%) had a haematologic response of PR or better, of which 42% were VGPR and above.[90]

This is further supported by data from the ALCHemy trial, a UK-based prospective study of patients with systemic AL amyloidosis, in which Stage III patients who did not respond to treatment had a survival rate of 20% at four years. The Stage III patients who did achieve a complete or very good partial haematologic response had an approximate survival rate of 75% at four years, underscoring the importance of achieving a deep and rapid haematologic response in this poor prognostic group in order to increase overall survival. Furthermore, given that haematologic responses are associated with improved survival and organ response, these data illustrate the reason that the primary aim of therapy is achievement and maintenance of this response.[91] Overall, these data indicate that receipt of, and response to, treatment has the potential to improve significantly the prognosis of these typically poor prognosis patients, highlighting the importance that these patients have access to DBCd should it be recommended by NICE.

It is therefore Janssen’s view that that any recommendation for DBCd in AL amyloidosis should not be restricted in such a way to exclude patients with Stage IIIb disease, a group of highly severe patients, who have an extremely poor prognosis and life expectancy and who have the potential to benefit greatly from new, effective treatment options.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

B.2 Clinical effectiveness

Summary of clinical effectiveness evidence

The ANDROMEDA trial

• The efficacy and safety of DBCd as compared with BCd was assessed in the ANDROMEDA trial: a randomised, open-label, multinational, multicentre Phase III trial of adult patients with newly diagnosed systemic AL amyloidosis.

• Expert clinical opinion confirms that the baseline demographic and disease characteristics of the enrolled population is broadly comparable to the AL amyloidosis population seen in clinical practice in England and Wales.[26]

• Data are presented from a pre-specified interim analysis (IA1; median follow-up 11.4 months) and a 12-month landmark analysis (median follow-up 20.3 months).

Haematologic response

• Results for haematologic response from ANDROMEDA demonstrate that DBCd achieves a deeper and more rapid haematologic response compared with the existing standard of care for AL amyloidosis in the UK, BCd.

• A statistically significantly higher rate of CHR was achieved in the DBCd treatment arm relative to the BCd arm. Additionally, DBCd induced at least a VGPR at a faster rate than BCd.

• The significant improvement in CHR rate in patients treated with DBCd was also observed across all pre-specified subgroups, including poor prognostic groups and importantly across patients with more advanced disease (Stage II and IIIa, according to the Mayo Clinical Staging System).

• Given the well-established relationship between depth of haematologic response and OS,[2, 27, 82, ] 92, 93 it is reasonable to expect that the deeper and more rapid haematologic responses achieved following DBCd treatment will translate into long-term improvements in OS for newly diagnosed AL amyloidosis patients.

• Achievement of CHR represents the optimal response in terms of patient prognosis and survival and is recommended as a key target in clinical treatment guidelines.[3-6]

Organ response

• Patients treated with DBCd achieved almost doubled rates of both cardiac and renal response at six months relative to those treated with BCd, with these significant improvements being sustained after 12- and 18-months.

• Furthermore, patients were able to achieve organ responses more quickly when treated with DBCd relative to BCd, delaying organ progression and enabling patients to avoid the detrimental impacts of organ deterioration on their quality of life. Results demonstrating an increased time to organ progression for patients treated with DBCd relative to BCd provide additional support to this.

• DBCd is therefore expected to fulfil a significant unmet need amongst AL amyloidosis patients, with many failing to achieve organ responses with existing bortezomib-based therapies.[27, 28, 30, ] 31

MOD-PFS and MOD-EFS

• The MOD-PFS endpoint allowed measurement of the time until patients experience significant progression of their disease, defined as the time from randomisation to one of the following events (whichever comes first): haematologic progression, end-stage cardiac or renal disease, or death.

• DBCd was shown to prolong the time to MOD-PFS relative to BCd, providing substantial benefits to patients in delaying serious deterioration of the heart and kidneys. This is expected to enable patients to avoid considerable negative impacts on their quality of life associated with end-stage organ failure.

• A further supplementary analysis of MOD-PFS, MOD-EFS, that incorporated patient switching to subsequent alternative therapies upon suboptimal response or worsening organ function as an event, further demonstrates the benefits of DBCd.

• The increased time to MOD-EFS events observed for DBCd as measured with this composite endpoint further highlights the ability of DBCd to increase the time during which AL amyloidosis

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

patients can avoid disease progression.

HRQoL

• Patients treated with DBCd in ANDROMEDA reported improved overall EQ-5D-5L utility scores. The results demonstrate that patients treated with DBCd experience improvements across important aspects of their quality of life as measured by the different scales in the EQ-5D-5L questionnaire.

• DBCd was associated with relatively stable EQ-5D-5L utility scores throughout the first six cycles of treatment, whilst patients in the BCd group reported a substantial decline during the same time period. From Cycle 7 onwards, as patients received subsequent daratumumab SC monotherapy, a consistent improvement in mean EQ-5D-5L utility scores was reported in the DBCd treatment group.

• Importantly, adding daratumumab SC to standard of care BCd combination therapy did not result in a detrimental effect on HRQoL, suggesting that DBCd may produce both improvements to clinical outcomes whilst at least maintaining patients’ HRQoL.

Safety

• The occurrence of treatment-emergent adverse events (TEAEs) and serious adverse events (SAEs, Grade III and IV) were broadly balanced across the treatment arms.

• TEAEs were generally manageable and did not lead to any increase in treatment discontinuation as compared with background therapy.

• Overall, DBCd was well tolerated, with a safety profile consistent with the established safety profiles of daratumumab SC and BCd and no new safety concerns were identified.

Conclusion

• In summary, the introduction of DBCd to UK clinical practice would provide a step-change in the care available for AL amyloidosis patients. Its use provides patients with a novel, highly effective therapeutic option with a tolerable safety profile which has the potential to improve patient prognosis and HRQoL, delay organ failure and prolong survival.

==> picture [32 x 12] intentionally omitted <==

Identification and selection of relevant studies

A de novo clinical systematic literature review (SLR) of the published literature was conducted to identify relevant clinical evidence (RCTs and non-RCTs) on the clinical efficacy and safety of pharmacological therapies for adults with newly diagnosed AL amyloidosis. Of note, the SLR took a broad approach and therefore included additional therapies not considered in the decision problem addressed in this submission.

The SLR was conducted according to a pre-specified protocol and performed in accordance with the methodological principles detailed in the PROSPERO international prospective register of systematic reviews.[94] The SLR was conducted in February 2021.

In total, the SLR identified five unique interventional studies (reported in eleven records) and 52 observational studies that met the inclusion criteria of the review. All five interventional studies were RCTs, and four of the five were Phase III trials.

Full details of the SLR search strategy, study selection process and results are presented in Appendix D.

==> picture [34 x 11] intentionally omitted <==

List of relevant clinical effectiveness evidence

Of the studies identified in the clinical SLR, the trial of direct relevance to the decision problem for this appraisal is ANDROMEDA (NCT03201965). ANDROMEDA is the pivotal registration trial, presented to the EMA in support of the marketing authorisation for daratumumab SC in Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

combination with BCd in adult patients with newly diagnosed systemic AL amyloidosis.[32] ANDROMEDA is a randomised, open-label, multinational, multicentre Phase III trial in patients at least 18 years of age with newly diagnosed systemic AL amyloidosis.[35] This study provides the main body of evidence for daratumumab SC in combination with the relevant comparator to this appraisal, BCd.

An overview of ANDROMEDA is presented in Table 7, and the methodology and results are presented in Section B.2.3 onwards.

Table 7: Clinical effectiveness evidence from ANDROMEDA

a Endpoints in bold are those that are used to inform the cost-effectiveness model. Abbreviations : AE: adverse event; AL: amyloid light-chain; BCd: bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; ECOG: Eastern Cooperative Oncology Group Performance Status; HRQoL: health-related quality of life; MOD-PFS: major organ deterioration-progression free survival; SC: subcutaneous. Source : Janssen ANDROMEDA Protocol.[35]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

==> picture [34 x 12] intentionally omitted <==

Summary of methodology of the relevant clinical

effectiveness evidence

==> picture [38 x 10] intentionally omitted <==

Trial design

ANDROMEDA is a randomised, open-label, multinational, multicentre Phase III trial, in patients at least 18 years of age, with newly diagnosed systemic AL amyloidosis.

Patients were randomised in a 1:1 ratio to receive either BCd alone (Treatment Arm A) or DBCd (Treatment Arm B), following stratification according to the following factors:[35]

• Cardiac stage based on the Mayo Clinical Cardiac Staging System (Stages I, II, and IIIa)

• Countries that typically offer (List A) or do not offer (List B) ASCT for patients with AL amyloidosis

• Renal function (creatinine clearance [CrCl] ≥60 mL/min versus CrCl <60 mL/min)

The trial design consisted of four phases, including a Screening Phase (extending up to 28 days prior to Cycle 1, Day 1), a Treatment Phase (from Cycle 1, Day 1 until study treatment discontinuation), a Post-Treatment Observation Phase and a Long-term Follow-up Phase.

A schematic of the study design of the ANDROMEDA trial is presented in Figure 3.

Figure 3: Design of the ANDROMEDA study

==> picture [490 x 87] intentionally omitted <==

==> picture [490 x 87] intentionally omitted <==

Each cycle is 28 days in length.

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; SC: subcutaneous; MOD-PFS: major organ deterioration-progression free survival. Source : Adapted from Kastritis et al. , (2020).[95]

During the Treatment Phase, patients in both the DBCd and BCd only arms received a maximum of six 28-day cycles of BCd therapy. Patients in the DBCd arm also received a fixed 1,800 mg dose of subcutaneous (SC) daratumumab, with weekly dosing during Cycles 1–2 and dosing every two weeks during Cycles 3–6. From Cycle 7 onwards, patients in the DBCd arm continued to receive daratumumab as monotherapy every four weeks until experiencing disease progression, starting a subsequent anti-plasma cell therapy, or a maximum of 24 cycles from the first dose of study treatment. An overview of the dosing schedule for the DBCd treatment arm is presented in Figure 4 below. A summary of the dosing schedule for the BCd arm is presented in Table 8 below.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Figure 4: Overview of ANDROMEDA daratumumab dosing schedule

==> picture [452 x 65] intentionally omitted <==

Abbreviations : SC: subcutaneous.

==> picture [39 x 11] intentionally omitted <==

Trial methodology

A summary of the methodology of ANDROMEDA is presented in Table 8 below.

Table 8: Summary of the ANDROMEDA trial methodology

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

 Time to cardiac response, time to renal response, and time to liver response: defined as the time between the date of randomisation and the first efficacy evaluation at which the patient had each corresponding organ response. Definitions of organ response can be found in Table 10  Duration of organ response: defined as the time between the date of initial documentation of each corresponding organ response to the date of first documented evidence of the corresponding organ progressive disease. For patients who did not have organ progression, data will be censored at the last disease assessment  Time to cardiac progression, time to renal progression, and time to liver progression: defined as the time from the date of randomisation to the date of each corresponding organ progression per consensus guidelines. Definitions of organ response can be found in Table 10  Improvement in fatigue: defined as the change from baseline in the EORTC QLQ-C30 Fatigue scale score, improvement in mental functioning is defined as the change from baseline in the SF-36v2 Mental Component Summary  Improvement in HRQoL: defined as change from baseline in the EORTC QLQ-C30 Global Health Status scale score  Mean EQ-5D-5L utility scores ranging from zero (0.0) to 1 (1.0) with higher values representing better general health status of the individual  Sex (male, female)  Age (<65, ≥65)  Baseline weight (≤65 kg, 65–85 kg, >85 kg)  Race (white, Asian, others)  Baseline cardiac stage (I, II, IIIa/b)  Transplant typically offered in local country (list A, list B) Pre-specified  Baseline renal function (≥60 mL/min, <50 mL/min) subgroups  Cardiac involvement at baseline (yes, no)  Baseline renal stage (I, II, III)  Baseline alkaline phosphatase (abnormal, normal)  Baseline ECOG performance status (0, 1 or 2)  Cytogenic risk at study entry (high risk, low risk)  FISH t(11;14) (abnormal, normal)

Abbreviations : AE: adverse event; AL: amyloid light-chain; BCd: bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; dFLC: uninvolved free light chains; ECOG: Eastern Cooperative Oncology Group Performance Status; EORTC QLQ-C30: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire; EQ-5D-5L: EuroQol-5 Dimensions-5 Level; FISH: florescence in situ hybridisation; HRQoL: health-related quality of life; IRC: Independent Review Committee; IV: intravenous; MM: multiple myeloma; MOD-PFS: major organ deterioration-progression free survival; NYHA: New York Heart Association; OrRR: organ response rate; OS: overall survival; SC: subcutaneous; SF-36v2: Short Form 36 Health Survey Questionnaire; UK: United Kingdom; VGPR: very good partial response. Source : Janssen ANDROMEDA Protocol.[35]

Rationale for choice of CHR as the primary endpoint

The primary endpoint used in ANDROMEDA was overall CHR, which was assessed via the consensus guidelines widely used in clinical practice in England to guide treatment options and assess patient prognosis.[24]

The rationale for selecting CHR as the primary endpoint in the trial was based primarily on its prognostic significance in relation to survival outcomes. The relationship between the timing and depth of haematologic response and improved OS is supported by a substantial body of evidence.[2, 27, 82, 92, 93] CHR therefore functions as a surrogate endpoint for survival, and was

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

selected in ANDROMEDA to enable modelling of survival in the absence of mature OS data from the trial.

Evaluation of haematologic response

The study protocol initially defined complete haematologic response as per Comenzo (2012) consensus guidelines criteria, which included achievement of a normalised free light chain (FLC) level and ratio, as well as negative serum and urine immunofixation.[24] However, this definition was later updated in line with subsequent publications that provided a broader understanding of the biological processes involved in AL amyloidosis.[18, 96, 97] The revised definition no longer required normalisation of the uninvolved FLC (uFLC) level and FLC ratio in patients who achieved an involved FLC (iFLC) level below the upper limit of normal (ULN). Disease evaluations were performed by a central laboratory.

Table 9 presents a summary of the criteria used to define the series of different haematologic response categories.

Rationale for choice of MOD-PFS as a major secondary endpoint

In clinical practice, disease progression in AL amyloidosis patients may be evaluated according to a range of biomarkers, including haematologic, cardiac and renal biomarkers given the heterogeneity in presentation of the disease. As a result of the complexities of using PFS to measure disease progression in AL amyloidosis, and to provide additional clinical context regarding the long-term durability of haematologic response and organ response, ANDROMEDA instead collected MOD-PFS. MOD-PFS is a novel, composite endpoint developed to encompass the most clinically relevant and objective measures of the benefits of anti-plasma cell therapy: haematologic progression, major organ deterioration, and death.

The use of MOD-PFS as a key secondary endpoint and measure of disease progression in the ANDROMEDA trial was approved following consultation with both the FDA and EMA.[33, 34]

Table 9: Summary of haematologic response endpoints and definitions based on the consensus guidelines

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

increase in urine M-protein to >200 mg/day (a visible peak must be present)  Involved free light chain increase of 50% to >100 mg/L

a Baseline measurement defined as the closest non-missing measurement taken on or prior to the first study treatment administration.

Abbreviations : CHR: complete haematologic response; dFLC: difference between the involved and uninvolved free light chain; FLC: free light chain; iFLC: involved free light chain; IRC: Independent Review Committee; NR: no response; PR: partial response; ULN; upper limit of normal; VGPR: very good partial response. Source : Comenzo et al ., (2012).[24]

Table 10: Summary of organ response and progression criteria based on the consensus guidelines

a Patients with progressive worsening renal function cannot be scored for NT-proBNP progression. Abbreviations : NT-proBNP: N-terminal prohormone of brain natriuretic peptide; cTnT: cardiac troponin; NYHA: New York Heart Association. Source : Comenzo et al ., (2012).[24]

==> picture [39 x 11] intentionally omitted <==

Baseline characteristics

Summaries of baseline demographic and disease characteristics of patients with AL amyloidosis included in ANDROMEDA are presented in Table 11 and Table 12, respectively. Overall, baseline demographics were well-balanced between the DBCd and BCd treatment arms. The median age was 64.0 years (range: 34–87), with 47.2% of the patients ≥65 years of age. Fiftyeight percent of patients were male. The majority of patients were white (75.8%) with an ECOG performance score of 0 (41.5%) or 1 (49.5%). Body weight subgroups were generally balanced between both treatment arms.[98]

Table 11: Baseline patient characteristics in ANDROMEDA (ITT population)

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations : BCd: bortezomib, cyclophosphamide, and dexamethasone DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide, and dexamethasone; ITT: intention-to-treat; SD: standard deviation. Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

Baseline disease characteristics were also well-balanced between the DBCd and BCd treatment arms. The median time since initial AL amyloidosis diagnosis to randomisation was 43.0 days. The median number of organs involved at baseline was 2, with 71.4% of patients having cardiac involvement, 59.0% of patients having kidney involvement, and 65.5% of patients having ≥2 organ involvement. 16.2% of patients were renal Stage III at baseline.[98]

Another key disease characteristic was Mayo Clinic Cardiac Stage at baseline. Disease staging in the ANDROMEDA trial was based primarily on the Mayo Clinic Staging systems described in Section B.1.3.1, but with some minor differences in the criteria used to categorise patients into stages. As compared with the criteria outlined in Table 6, the ANDROMEDA trial implemented an hs-cTnT threshold of 54 ng/L instead of a cTnT threshold of 0.035 µg/mL, and Stage III patients were divided into IIIa and IIIb based on the NT-proBNP threshold of 8,500 ng/L alone without consideration of a systemic blood pressure factor (threshold of 100 mg Hg) that was used to further divide Stage III patients in the 2013 revision of the Mayo staging system.[16] Approximately Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

one-third (36.6%) of patients were Mayo Clinic Cardiac Stage IIIa/b at baseline. It should be noted that although patients with Stage IIIb were excluded from ANDROMEDA, eight patients (two and six in the DBCd and BCd arms, respectively) who did not have Stage IIIb disease at screening, progressed to Stage IIIb at the time of first study dose administration.

Table 12: Baseline patient disease characteristics in ANDROMEDA (ITT population)

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

a Cardiac stage is based on both NT-proBNP and hs-cTnT levels; b Renal stage is based on eGFR and proteinuria testing;[c] Chronic kidney disease stage is based on eGFR;[d] Cytogenetic risk is based on FISH or karyotype testing. High risk is defined as: 1) by FISH testing: t (4; 14), t(14; 16), and 17p deletion; or 2) by Karyotype testing: t (4; 14), 17p deletion.

Abbreviations : ANS: autonomic nervous system; BCd: bortezomib, cyclophosphamide, and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide, and dexamethasone; dFLC: difference in involved and uninvolved free light chains; ECOG: Eastern Cooperative Oncology Group; eGFR: estimated glomerular filtration rate; FISH: florescence in situ hybridization; FLC: free light chain; iFLC: involved free light chain; hs-cTnT high sensitivity cardiac troponin T; ITT: intention-to-treat; NT-proNBP: N-terminal pro b- type natriuretic peptide; NYHA: New York Heart Association; PNS: peripheral nervous system; SD: standard deviation.

Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

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

==> picture [38 x 10] intentionally omitted <==

Trial populations

The definitions of the ANDROMEDA study populations are presented in Table 13.

Table 13: Trial populations used for the analysis of endpoints of ANDROMEDA

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations : DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide, and dexamethasone; ITT: intention-to-treat; NT-proNBP: N-terminal pro b-type natriuretic peptide; NYHA: New York Heart Association; rHuPH20: recombinant human hyaluronidase PH20. Source : Janssen ANDROMEDA Protocol[98]

==> picture [39 x 10] intentionally omitted <==

Statistical methods

The statistical analyses used to analyse the primary endpoint (overall CHR rate as assessed by blinded IRC), alongside sample size calculations and methods for handling missing data, are presented in Table 14.

Table 14: Statistical methods for the primary analysis of ANDROMEDA

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

© Janssen-Cilag (2021). All rights reserved

==> picture [79 x 100] intentionally omitted <==

withdrawal was to be documented in the electronic case report form and in the source document. Study drug assigned to the withdrawn patient may not be assigned to another patient. Patients who withdrew from the study were not replaced  The patient could withdraw consent for use of samples for research. In such a case, samples were destroyed after they were no longer needed for the clinical study  See Table 8 for details on censoring of missing data for outcomes

Abbreviations : AL: amyloid light-chain; BCd: bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; MOD-PFS: major organ deterioration-progression free survival; OS: overall survival. Source : Janssen ANDROMEDA Protocol.[35]

==> picture [39 x 10] intentionally omitted <==

Participant flow in the relevant randomised controlled trials

A summary of patient flow in ANDROMEDA is presented in Appendix D.

==> picture [34 x 12] intentionally omitted <==

Quality assessment of the relevant clinical effectiveness

evidence

RCTs captured in the clinical SLR were assessed for quality using the NICE clinical effectiveness quality assessment checklist, and non-RCTs and observational studies were assessed using the Newcastle-Ottawa scale. The results of these quality assessments are presented in Appendix D, and a summary of the quality assessment for ANDROMEDA is presented in Table 15 below.

Table 15: Quality assessment of the ANDROMEDA trial

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; ITT: intention-to-treat.

==> picture [34 x 12] intentionally omitted <==

Clinical effectiveness results of the relevant trials

Efficacy results from ANDROMEDA in this submission are presented from a pre-specified interim analysis (IA1; data cut-off 14[th] February 2020) and a 12-month landmark analysis (data cut-off 13[th] November 2020) and are based on a blinded IRC assessment of the outcomes investigated in the ANDROMEDA trial. The 12-month landmark analysis was not a pre-specified data cut, and instead was generated for conference purposes. The next pre-specified interim analysis for MOD-PFS and OS will occur when 200 MOD-PFS events have been observed in ANDROMEDA. Where possible, efficacy results for outcomes assessed in both the pre-specified interim analysis and 12-month landmark analysis are presented in parallel. For a number of outcomes, data are only available from the pre-specified interim analysis.

A summary of the outcomes from the pre-specified interim analysis and 12-month landmark analysis that are presented in the submission is provided in Table 16.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 16: Summary of ANDROMEDA data cuts

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved Page 51 of 165

==> picture [68 x 66] intentionally omitted <==

Abbreviations: CHR: complete haematologic response; EORTC QLQ-C30: European Organization for Research and Treatment of Cancer Quality of Life Questionnaire Version 3.0; EQ-5D-5L: EuroQol-5 Dimensions-5 Level; GHS: Global Health Status; ITT: intention-to-treat; MCS: Mental Component Summary; MOD-EFS: major organ deterioration-event free survival; MOD-PFS: major organ deterioration-progression free survival; MRD: minimal residual disease; OS: overall survival; PFS: progression-free survival; SF-36v2: Short Form 36 Health Survey Questionnaire; VGPR: very good partial response.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

==> picture [38 x 10] intentionally omitted <==

Haematologic response

As described in Section B.1.3.2, the primary therapeutic goal in the treatment of AL amyloidosis is to achieve a rapid, deep and durable haematologic response.[24, 25] This is a clinically meaningful outcome, as multiple studies have established the relationship between deeper haematologic response and improved prognosis for AL amyloidosis patients, with each successive category of response achieved associated with delayed disease progression, improved organ response rates and overall survival.[2, 27, 82] Haematologic response comprises a key goal in clinical guidelines for AL amyloidosis, which recommend that clinicians treat patients to target VGPR as a minimum, with CHR representing the optimal response in terms of patient prognosis and survival.[3-6]

Primary efficacy analysis

At a median follow-up duration of 11.4 months, as assessed by blinded IRC, the addition of daratumumab SC to BCd resulted in a statistically significant and clinically meaningful improvement in the overall CHR rate compared with BCd alone (53.3% vs 18.1%, respectively; odds ratio [OR]: 5.13; 95% CI: 3.22, 8.16; p<0.0001; Table 17). In the 12-month landmark analysis, with a median follow-up duration of 20.3 months, DBCd continued to give rise to a deeper haematologic response than BCd alone. A significantly greater proportion of patients achieved CHR in the DBCd group compared with BCd alone (59.0% vs 19.2%, respectively; OR: 5.90; 95% CI: 3.72, 9.37; p<0.0001; Table 17).

At a median follow-up duration of 11.4 months, the proportion of patients achieving VGPR or better was also statistically significant and clinically superior for DBCd compared with BCd alone **** **** (78.5% vs 49.2%, respectively; OR: 3.75; 95% CI: , ; p<0.0001; Table 17). Similarly, in the 12-month landmark analysis, the proportion of patients achieving VGPR or better remained significantly greater in the DBCd group than in the BCd group (79.0% vs 50.3%; OR: 3.74; 95% CI: 2.39, 5.86; p<0.0001; Table 17).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 17: Summary of overall best confirmed haematologic response based on IRC assessment; ITT analysis set (14[th ] February 2020 data cut-off and 13[th] November 2020 data cut-off)

a 95% CIs are based on Clopper-Pearson exact test. b Mantel-Haenszel estimate of the common odds ratio for stratified tables is used. The stratification factors from IWRS are: cardiac staging (I, II, IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A, List B), and baseline renal function (CrCl ≥60 mL/min or CrCl <60 mL/min). An odds ratio > 1 indicates an advantage for DBCd.[c] P-value from the Cochran Mantel-Haenszel Chi-Squared test. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; ITT: intention-to-treat; NE: not evaluable; NR: no response; PD: progressive disease; VGPR: very good partial response. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020);[95] Janssen ANDROMEDA 12-month landmark analysis (2021);[99] Kastritis et al., (2021).[100]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Six- and 12-month CHR rates

A major secondary outcome was achievement of CHR at 6 and 12 months. At the pre-specified interim analysis (IA1), greater six- and 12-month CHR rates were observed in the DBCd arm compared with the BCd arm (six months: ***** vs *****, respectively [OR: ****; 95% CI: ****, *****]; 12 months: ***** vs ****, respectively [OR: ****; 95% CI: ****, ****]).

In the 12-month landmark analysis, the greater confirmed CHR rate at six months in the DBCd group compared with the BCd group was maintained. Results for the CHR rate at 12 months were not available from the 12-month landmark analysis.

Results for achievement of CHR at 6 and 12 months for the IA1 analysis and achievement of CHR at six months for the 12-month landmark analysis are reported in Table 18. Due to the median follow-up duration of 11.4 months at the IA1 analysis, a lower CHR rate at 12 months was anticipated, given the relatively high proportion of patients who had not yet reached 12 months of follow-up. The apparent reduction in CHR rate between the 6- and 12-month timepoint can therefore be explained by this short follow-up, rather than by loss of response.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 18: Summary of confirmed CHR at six- and 12-months based on IRC assessment, ITT analysis set (14[th] February 2020 data cut-off and 13[th] November 2020 data cut-off)

a 95% CIs are based on Clopper-Pearson exact test. b Mantel-Haenszel estimate of the common odds ratio for stratified tables is used. The stratification factors from IWRS are: cardiac staging (I, II, IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A, List B), and baseline renal function (CrCl ≥ 60 mL/min or CrCl <60 mL/min). An odds ratio > 1 indicates an advantage for DBCd.[c] P-value from the Cochran Mantel-Haenszel Chi-Squared test. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; CHR: complete haematologic response; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC: independent review committee; NR: not reported. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020);[95] Janssen ANDROMEDA 12-month landmark analysis (2021).[99]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Time to haematologic response

A major secondary endpoint was time to haematologic response. As discussed previously, early and profound reductions of amyloid light chains are associated with the greatest chance of organ improvement, delayed progression and prolonged overall survival.[72]

Among patients who achieved CHR at the pre-specified interim analysis (IA1), median time to CHR was 60 days in the DBCd arm and 85 days in the BCd arm. Among subjects who achieved VGPR or better, median time to VGPR was 17 days in the DBCd arm and 25 days in the BCd arm (Table 19).

In the 12-month landmark analysis, among patients who achieved CHR, median time to CHR was 2.04 months (approximately 62 days) in the DBCd group and 2.79 months (approximately 85 days) in the BCd group. Median time to VGPR or better was also shorter in the DBCd group (0.56 months; approximately 17 days) compared to the BCd group (0.82 months; approximately 25 days).

Results for time to haematologic response for the IA1 and 12-month landmark analyses are presented in Table 19.

Table 19: Summary of time to haematologic response based on IRC assessment; haematologic response-evaluable analysis set (14[th] February 2020 data cut-off and 13[th] November 2020 data cut-off)

VGPR or better includes CR and VGPR. PR or better includes CR, VGPR and PR. Hematologic responseevaluable set includes subjects who have a confirmed diagnosis of amyloidosis and measurable disease at baseline or screening visit. In addition, subjects must have received at least 1 administration of study treatment and have at least 1 post- baseline disease assessment.[a] Time from randomisation date up to the first response of complete hematologic response is summarised.[b] Time from randomisation date up to the first response of VGPR or better, whichever is the earliest, is summarised.[c] Time from randomisation date up to the first response of PR or better, whichever is the earliest, is summarised.

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; CHR: complete haematologic response; IRC: independent review committee; SD: standard deviation; PR: partial response VGPR: very good partial response.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020);[95] Janssen ANDROMEDA 12-month landmark analysis (2021);[99] Kastritis et al., (2021).[100]

Results from both the IA1 and 12-month landmark analysis demonstrate that DBCd is able to give rise to a more rapid haematologic response when compared with BCd, which in turn is expected to delay deterioration of organ condition.

Duration of haematologic response

Interim analysis (IA1): Data cut-off 14[th] February 2020

Another major secondary endpoint was the duration of haematologic response. A prolonged haematological response is critical for delaying disease progression and improving survival.

At the pre-specified interim analysis (IA1), with a median follow-up of 11.4 months, the median duration of CHR had not been reached in either treatment arm (range: ***** ** ***** months for DBCd; ***** ** ***** months for BCd). Of the *** patients who achieved CHR in the DBCd arm, * patients died while in CHR and * patients relapsed following CHR. Of the ** patients who achieved CHR in the BCd arm, * died while in CHR and * patients relapsed following CHR (Table 20).

Table 20: Summary of duration of CHR based on IRC assessment; responders in ITT analysis set (14[th] February 2020 data cut-off)

a Events are defined as disease relapses, with deaths not counted as events. Percentages are calculated with the number of subjects in each treatment group with available data as denominator. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; NE: not estimable. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off)[98]

Results for the duration of CHR are not available from the 12-month landmark analysis.

In summary, results for haematologic response from ANDROMEDA demonstrate that DBCd may give rise to a deeper and more rapid haematologic response compared to the existing standard of care for AL amyloidosis in the UK, BCd. This is a clinically meaningful outcome, as multiple studies have established the relationship between deeper haematologic response and improved prognosis for AL amyloidosis patients, with each successive category of response achieved associated with delayed disease progression, improved organ response rates and overall survival.[2, 27, 82]

==> picture [39 x 10] intentionally omitted <==

Major organ deterioration progression-free survival (MOD-PFS)

As described in Section B.2.3.2, MOD-PFS is a composite endpoint of multiple clinically observable endpoints, defined as the time from patient randomisation to either clinical Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

manifestation of either cardiac or renal failure, development of haematologic progressive disease as per Comenzo et al., (2012) consensus guidelines, or death (whichever comes first).[24]

Disease progression in AL amyloidosis is evaluated according to a range of different biomarkers in clinical practice (because of the heterogeneity in presentation of disease). Due to the complexity in defining PFS in AL amyloidosis, ANDROMEDA collected MOD-PFS, and use of this endpoint has been approved by the FDA and EMA as a clinically relevant measure of both disease and the benefits of anti-plasma cell therapy.[33, 34]

The treatment paradigm used in ANDROMEDA involved patients being switched to an alternative treatment following a suboptimal haematologic response or worsening organ function, which commonly occurs prior to disease progression in clinical practice. As a result, this may have interfered with evaluation of the MOD-PFS endpoint, for which haematologic progression is among the outcomes included in the composite endpoint. Therefore, the primary analysis for MOD-PFS in ANDROMEDA employed the inverse probability of censoring weighting (IPCW) method to adjust estimates of the treatment effect in the presence of subsequent non-cross resistant, anti-plasma cell therapy. Full statistical details of the IPCW analysis can be found in Appendix L.

To test the robustness of the primary analysis results, pre-specified sensitivity IPCW analyses which used a different variable selection modelling and weight calculation approach were also conducted. Pre-specified sensitivity analyses of MOD-PFS were also performed. This included naïve censoring of subsequent non-cross resistant, anti-plasma cell therapy. In addition, given that patients would be permitted to switch in clinical practice if they do not achieve an adequate response, a supplementary analysis of MOD-PFS based on IRC assessment without censoring for any subsequent non-cross resistant, anti-plasma cell therapy, was also conducted.

Interim analysis (IA1): Data cut-off 14[th] February 2020

At a median follow-up duration of 11.4 months, after adjusting for dependent censoring due to switching to subsequent non-cross resistant, anti-plasma cell therapy, a substantial improvement in MOD-PFS was observed for patients receiving DBCd compared with BCd alone. The hazard ratio for MOD-PFS for DBCd vs BCd based on the primary IPCW analysis was ***** (95% CI: ***** ***** , ); this indicates a reduced risk of experiencing a MOD-PFS event in the DBCd arm compared to the BCd arm. Median MOD-PFS was not reached in either treatment arm. While the ****** nominal p-value for this interim analysis was , above the pre-specified significance threshold of *******, there is a clear, substantial difference in treatment effect between DBCd and BCd, as demonstrated by the clear separation of the two Kaplan-Meier curves after Month 6 (Figure 5).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 5: Kaplan-Meier plot of MOD-PFS based on IRC assessment, IPCW analysis; ITT analysis set (14[th] February 2020 data cut-off)

==> picture [452 x 291] intentionally omitted <==

Abbreviations: CI: confidence interval; CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); Dara SC: daratumumab subcutaneous. Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

The results from pre-specified sensitivity IPCW analyses, which used a different variable selection modelling and weight calculation approach, were consistent with the primary IPCW analysis. Further pre-specified sensitivity analyses of MOD-PFS such as naïve censoring of subsequent non-cross resistant, anti-plasma cell therapy were also performed, and a supplementary analysis of MOD-PFS based on IRC assessment without censoring for any subsequent non-cross resistant, anti-plasma cell therapy also demonstrated consistency with the results from the primary analysis (Table 21).

Table 21: Summary of primary, sensitivity and supplementary analysis of MOD-PFS based on IRC assessment; ITT analysis set (14[th] February 2020 data cut-off)

a Refers to subsequent non-cross resistant, anti-plasma cell therapy.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC: independent review committee; HR: hazard ratio; IPCW: inverse probability of censoring weight; MOD-EFS: major organ deterioration-event-free survival; MOD-PFS: major organ deterioration-progression-free survival. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Updated MOD-PFS results are not available from the 12-month landmark analysis.

Overall, the MOD-PFS results from the pre-specified interim analysis were robust and consistent, favouring the DBCd arm and demonstrating a substantial delay in haematologic progression, major organ deterioration, or death.

Increasing the length of time to major organ deterioration offers significant value for patients, given the very poor prognosis associated with progression of cardiac and renal disease to later stages.[20, 70] The development of end-stage organ failure is likely to have substantial negative impacts on patient quality of life, with an increasing burden of severe disease symptoms, increased frequency of hospital visits and the continuation of poorly tolerated chemotherapy treatments. Further, progression of AL amyloidosis to end-stage organ failure results in an increased burden to the NHS, such as the significant costs of dialysis to manage end-stage renal failure.[22, 23]

Major organ deterioration event-free survival (MOD-EFS)

As discussed above, in ANDROMEDA, patients were able to switch to subsequent non-cross resistant, anti-plasma cell therapy before haematologic progression or major organ deterioration in cases where they experienced suboptimal haematologic response or worsening organ function. Initiation of subsequent therapy therefore represents a key measure of both the rate and depth of haematologic response, as patients with delayed or suboptimal response may need to switch to a subsequent therapy.

As a switch to subsequent therapy is not captured by MOD-PFS assessments, a supplementary analysis of MOD-PFS was explored which included subsequent non-cross resistant, anti-plasma cell therapy as an event. MOD-EFS is therefore a composite endpoint incorporating haematologic progression, major organ deterioration, initiation of any subsequent non-cross resistant anti-plasma cell therapy, or death, whichever event comes first.

Interim analysis (IA1): Data cut-off 14[th] February 2020

In the pre-specified interim analysis, the median MOD-EFS was *** months in BCd treatment arm, and was not yet reached in the DBCd arm (HR: ****; 95% CI: ****, ****; nominal p-value: *******; Figure 6). The hazard ratio indicates that there is a reduced risk of a MOD-EFS event in the DBCd group compared with the BCd group.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 6: Weighted Kaplan-Meier plot of MOD-EFS based on IRC assessment; ITT analysis set (14[th] February 2020 data cut-off)

==> picture [373 x 133] intentionally omitted <==

==> picture [373 x 133] intentionally omitted <==

==> picture [373 x 133] intentionally omitted <==

Abbreviations: CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); Dara SC: daratumumab subcutaneous; IRC: independent review centre; MOD-EFS: major organ deterioration eventfree survival. Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

Updated MOD-EFS results are not available from the 12-month landmark analysis.

==> picture [39 x 11] intentionally omitted <==

Overall survival

Interim analysis (IA1): Data cut-off 14[th] February 2020

OS data were not mature at the time of the pre-specified interim analysis (IA1), with 56 deaths in total (27 in the DBCd arm and 29 in the BCd arm), including 1 randomised subject in the BCd arm who died without receiving study treatment (Table 22; Figure 7). The HR for survival was **** (DBCd vs BCd; 95% CI: ****, ****), and the nominal p-value was ******. Median OS was not reached in either treatment arm, and the estimated 18-month OS was ***** in the DBCd arm and ***** in the BCd arm.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Table 22: Summary of OS; ITT analysis set (14[th] February 2020 data cut-off)

a P-value is based on a log-rank test stratified with cardiac stage (Stage I, II, and IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A or List B), and renal function (CrCl ≥ 60 mL/min or CrCl <60 mL/min) as randomised.[b] Hazard ratio and 95% CI from a Cox proportional hazards model with treatment as the sole explanatory variable and stratified with cardiac stage (Stage I, II, and IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A or List B), and renal function (CrCl ≥60 mL/min or CrCl <60 mL/min) as randomised. A hazard ratio <1 indicates an advantage for DBCd. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; OS: overall survival; NE: not evaluable.

Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Figure 7: Kaplan-Meier plot for OS; ITT analysis set (14[th] February 2020 data cut-off)

==> picture [383 x 337] intentionally omitted <==

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations: CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); Dara SC: daratumumab subcutaneous.

Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Updated OS results are not available from the 12-month landmark analysis.

OS data from ANDROMEDA remain immature at currently available data cut offs. However, given the established relationship between haematological response and overall survival in AL amyloidosis,[2, 27, 82] results presented in Section B.2.6.1 illustrating the rapid and deep haematological response achieved with DBCd compared to BCd are expected to result in longterm survival benefits for patients treated with DBCd.

==> picture [39 x 11] intentionally omitted <==

Time to subsequent non-cross resistant anti-plasma cell therapy

In line with the treatment paradigm for patients with AL amyloidosis and with the ANDROMEDA protocol, patients with suboptimal haematologic response or worsening organ function were permitted to start subsequent non-cross resistant, anti-plasma cell therapy prior to developing haematologic progression after three cycles of treatment.

Non-cross resistant anti-plasma cell therapy was defined as ASCT with high dose melphalan, melphalan plus dexamethasone, or any new combination regimen that included at least one new component that was different to the assigned study drugs received (i.e. bortezomib plus lenalidomide for both treatment arms and daratumumab SC for the BCd arm).

A summary of the proportion of patients that received different subsequent therapies in ANDROMEDA is presented in Table 23 below. More patients in the BCd arm (** patients [] received subsequent therapy (both cross-resistant and non-cross resistant), compared with those in the DBCd arm ( patients []. Of those patients who received subsequent therapy, *** (/) patients in the BCd arm and *** (/) in the DBCd arm received therapy that met the criteria for non-cross resistant subsequent therapy, in line with the definition above. The most common non-cross resistant subsequent therapy in the DBCd arm was melphalan (/*** [****], and the most common therapy in the BCd arm was daratumumab IV (48/188 [*****]).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 23: Summary of subsequent non-cross resistant, anti-plasma cell therapy by therapeutic class, pharmacologic class, and preferred term; safety analysis set (14[th] February 2020 data cut-off)

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone. Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

A summary of the time to subsequent non-cross resistant, anti-plasma cell therapy for both the pre-specified interim analysis and 12-month landmark analysis is presented in Table 24.

In the pre-specified interim analysis, more patients in the BCd arm () received subsequent non-cross resistant anti-plasma cell therapy compared with patients in the DBCd arm (), whilst in the 12-month landmark analysis, the proportion of patients receiving non-cross resistant anti-plasma cell therapy remained higher in the BCd arm () compared with the DBCd arm () (Table 24).

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

In the pre-specified interim analysis, the median time to initiation of subsequent non-cross resistant anti-plasma cell therapy was not reached for subjects in the DBCd arm, and was 10.38 months in the BCd arm (HR: ****; 95% CI: ****, ****; ********; Table 24). In the 12-month landmark analysis, the median time to subsequent non-cross resistant anti-plasma cell therapy was still yet to be reached, whilst it was ***** ****** (95% CI: ****, *****) in the BCd arm (Table 24).

Kaplan-Meier curves for the time to first subsequent non-cross resistant anti-plasma cell therapy for both the pre-specified interim analysis (Figure 8) and 12-month landmark analysis (Figure 9) are also presented below. Separation of the Kaplan-Meier curves is observed after Month 3 in the case of both the IA1 and 12-month landmark analyses, with a clear treatment effect between DBCd and BCd observable at Month 6 (Figure 8 and Figure 9).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 24: Summary of time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (14[th] February 2020 and 13[th] November 2020 data cut-off)

a p-value is based on a log-rank test stratified with cardiac stage (Stage I, II, and IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A or List B), and renal function (CrCl ≥60 mL/min or CrCl <60 mL/min) as randomised.[b] Hazard ratio and 95% CI from a Cox proportional hazards model with treatment as the sole explanatory variable and stratified with cardiac stage (Stage I, II, and IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A or List B), and renal function (CrCl ≥60 mL/min or CrCl <60 mL/min) as randomised. A hazard ratio <1 indicates an advantage for DBCd. Abbreviations : BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; ITT: intention-to-treat; NE: not estimable. Source : Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Janssen ANDROMEDA 12-month landmark analysis (2021).[99]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 8: Kaplan-Meier plot for time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (14[th] February 2020 data cut-off)

==> picture [391 x 348] intentionally omitted <==

Abbreviations: CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); Dara SC: daratumumab subcutaneous; ITT: intention-to-treat. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 9: Kaplan-Meier plot for time to first subsequent non-cross resistant anti-plasma cell therapy; ITT analysis set (13[th] November 2020 data cut-off)

==> picture [334 x 337] intentionally omitted <==

Abbreviations: D-VCd: daratumumab in combination with bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as DBCd); ITT: intention-to-treat; VCd: bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as BCd).

Source: ANDROMEDA 12-month landmark analysis (November 2020).[99]

Overall, the time to initiation of subsequent non-cross resistant, anti-plasma cell therapy represents an additional measure of both the rate and depth of haematologic response, given that patients with suboptimal response or worsening of organ function may be switched onto subsequent therapies as early as after three cycles. As compared with BCd, treatment with DBCd increased the time to initiation of subsequent therapies.

==> picture [39 x 10] intentionally omitted <==

Cardiac, renal and liver responses

In AL amyloidosis, the systemic nature of the disease results in amyloid deposition in organs throughout the body which can substantially impair organ function and may ultimately lead to organ failure.[45, 101, 102] Cardiac failure can lead to death, while renal failure can mean patients require renal replacement therapy and has significant impacts on patient quality of life, alongside greatly increasing costs to the NHS.[22, 23] Organ response therefore represents a key outcome in the ANDROMEDA trial to understand the relative effectiveness of DBCd vs BCd alone for delaying the amyloid deposition in organs which may ultimately lead to organ failure.

As was the case for evaluation of MOD-PFS, the option for patients to switch treatment to noncross resistant, anti-plasma cell therapy in cases of suboptimal haematologic response or worsening organ function meant that analysis of organ response may have been affected by the treatment patients subsequently went on to receive. It was therefore necessary to conduct

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

analyses of organ responses both with and without censoring for subsequent non-cross resistant, anti-plasma cell therapy, to understand the effect of DBCd and BCd on organ response rates.

Organ response rate

Interim analysis (IA1): Data cut-off 14[th] February 2020

Of the *** patients with baseline cardiac involvement, *** were evaluable for cardiac response (DBCd: n=118; BCd: n=117). A substantially greater cardiac response was observed in the DBCd arm compared to the BCd arm; cardiac response rate at six months for patients in the DBCd arm was nearly twice that of patients in the BCd arm (41.5% vs 22.2%; OR: ****; 95% CI: **** **** , ; p=0.0029; without censoring for subsequent non-cross resistant anti-plasma cell therapy). Results were consistent regardless of whether the analysis was conducted with or without censoring for subsequent non-cross resistant anti-plasma cell therapy.

There were *** patients evaluable for renal response (DBCd: 113; BCd: 117). Similar to cardiac response, a substantially greater renal response was observed in the DBCd arm compared to the BCd arm. The renal response rate at six months was 53.8% in the DBCd arm compared with 27.4% in the BCd arm (OR: ****; 95% CI: ****, ****; p<0.0001; without censoring for subsequent non-cross resistant anti-plasma cell therapy). Again, the results were comparable regardless of whether the analysis was conducted with or without censoring for subsequent non-cross resistant anti-plasma cell therapy.

The liver response rate at six months was substantially higher in the DBCd arm compared with the BCd arm (***** vs ****, respectively), both with and without censoring for subsequent noncross resistant anti-plasma cell therapy. However, it is not possible to make definitive comparative conclusions with regards to liver response rates due to the limited number of ** ** ** evaluable patients ( ; DBCd: n= ; BCd: n= ).

Cardiac and renal response rates at six months for the pre-specified interim analysis are presented in Table 25.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 25: Cardiac and renal six-month response rates (14[th] February 2020 data cut-off)

a Cardiac response was based on NT-proBNP response (>30% and >300 ng/L decrease in subjects with baseline NT-proBNP >650 ng/L) or NYHA class response (>2 class decrease in subjects with baseline NYHA class 3 or 4) per Comenzo (2012) consensus criteria.[24][b] Renal response was defined as ≥30% decrease in proteinuria or proteinuria decreased to <0.5 g/24 hours in the absence of renal progression. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC: independent review committee. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

12-month landmark analysis: Data cut-off 13[th] November 2020

In the 12-month landmark analysis, updated results for organ response rates at six months are presented, along with organ response rates at the 12- and 18-month timepoints. In this later analysis timepoint, only results without censoring for subsequent non-cross resistant, anti-plasma cell therapy were available.

Among the *** cardiac response-evaluable patients with cardiac involvement at baseline (DBCd: n=118; BCd: n=117), similarly to the interim analysis, cardiac response rates were substantially higher with DBCd arm compared to BCd. DBCd was associated with approximately ******** ****** rates of cardiac response than BCd at 6, 12 and 18 months (****** for all comparisons; Table 26). Compared to the interim analysis, organ response rates after longer follow-up were greater.

Similarly, among the renal response-evaluable patients with renal involvement at baseline (DBCd: n=117; BCd: n=113), DBCd was associated with approximately ******** ****** rates of renal response at 6, 12 and 18 months (******* for all comparisons; Table 27).

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Table 26: Summary of cardiac response rate at 6, 12 and 18 months based on IRC assessment without censoring non-cross resistant anti-plasma cell therapy; cardiac response-evaluable analysis set (13[th] November 2020 data cut-off)

Cardiac response is determined by evaluation of NYHA and NT-proBNP decrease from baseline.[a] 95% CIs are based on Clopper-Pearson exact test.[b] Mantel-Haenszel estimate of the common odds ratio for stratified tables is used. The stratification factors from IWRS are: cardiac staging (I, II, IIIa), countries that typically offer or not offer transplant for patients with AL amyloidosis (List A, List B), and baseline renal function (CrCl ≥60 mL/min or CrCl <60 mL/min). An odds ratio > 1 indicates an advantage for DBCd.[c] P-value from the Cochran MantelHaenszel Chi-Squared test. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd; daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC; independent review committee.

Source: Janssen ANDROMEDA 12-month landmark analysis (2021);[99] Kastritis et al., (2021).[100]

Table 27: Summary of renal response rate at 6, 12 and 18 months based on IRC assessment without censoring non-cross resistant anti-plasma cell therapy; renal response-evaluable analysis set (13[th] November 2020 data cut-off)

a 95% CIs are based on Clopper-Pearson exact test. b Mantel-Haenszel estimate of the common odds ratio for stratified tables is used. The stratification factors from IWRS are: cardiac staging (I, II, IIIa), Countries that typically offer or not offer transplant for patients with AL amyloidosis (List A, List B), and baseline renal function (CrCl ≥60 mL/min or CrCl <60 mL/min). An odds ratio > 1 indicates an advantage for DBCd.[c] P-value from the Cochran Mantel-Haenszel Chi-Squared test. Renal response indicates ≥30% decrease in proteinuria or drop in proteinuria below 0.5 g/24 hour in the absence or renal progression.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd; daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC; independent review committee.

Source: Janssen ANDROMEDA 12-month landmark analysis (2021);[99] Kastritis et al., (2021).[100]

Time to organ response

Interim analysis (IA1): Data cut-off 14[th] February 2020

The median times to cardiac, renal and liver responses as per IRC assessment are presented in Table 28. Cardiac and renal responses were reached more quickly in the DBCd group than in the BCd group, and this result was observed both with and without censoring for non-cross resistant anti-plasma cell therapy. The median time to cardiac response was **** months in the DBCd group and **** months in the BCd group, with censoring for subsequent therapy. The median time to renal response was also reached approximately one month earlier in the DBCd group, at **** months compared with **** months in the BCd group. The time to liver response was faster in the DBCd group without censoring for subsequent therapy, though the smaller sample size of evaluable patients precludes meaningful comparisons.

Table 28: Median time to cardiac, renal and liver response based on IRC assessment (14[th] February 2020 data cut-off)

a The median time to organ response is reported for evaluable responding patients in the DBCd group (cardiac n=59; renal n=83; liver n=5) and in the BCd group (cardiac n=41; renal n=45; liver n=2). Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd; daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; IRC; independent review committee. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Cardiac, renal and liver progression rates at six months

Interim analysis (IA1): Data cut-off 14[th] February 2020

A summary of the proportion of evaluable patients with cardiac, renal and liver progression after six months is presented in Table 29.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Compared with the BCd group, numerically lower rates of cardiac, renal and liver progression were observed after six months among evaluable patients with organ involvement at baseline. At the six-month time point, 13.6% of evaluable patients in the DBCd group had experienced cardiac progression, compared with 19.7% in the BCd group.

Table 29: Cardiac, renal and liver progression rates at six months based on IRC assessment (14[th] February 2020 data cut-off)

a The median time to organ response is reported for evaluable responding patients in the DBCd group (cardiac n = 59; renal n = 83; liver n = 5) and in the BCd group (cardiac n=41; renal n=45; liver n=2). Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CI: confidence interval; DBCd; daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Overall, DBCd resulted in a clinically meaningful, substantial improvement in cardiac and renal response rates when compared with BCd, with a nearly doubled rate of improvement at six months. These improvements, in combination with the increased rate of response, again provide a substantial benefit to patients in the context of the generally poor organ response rates achieved by those receiving treatment with existing bortezomib-based therapies.[27, 28, 30, 31] Greater organ response rates and delayed time to organ progression may also mean that patients avoid some of the substantial impacts that AL amyloidosis symptoms can have on their ability to carry out their daily lives, particularly those symptoms related to cardiac involvement.[76] 77

Similarly, these improvements to organ response rates would be expected to delay disease progression to late-stage organ failure, where the impacts for patients can be very severe. In the case of cardiac failure this can lead to death, while end-stage renal failure may require patients to begin dialysis or receive a kidney transplant. [18, 19, 57, 60]

Furthermore, the fact that the near two-fold improvements in organ response rate for DBCd compared with BCd observed in the pre-specified interim analysis were sustained in the 12month landmark analysis suggests that the benefits to patients described may also be sustained in the short to medium term.

==> picture [39 x 10] intentionally omitted <==

Health-related quality of life

In the ANDROMEDA trial, data from a series of health-related quality of life instruments were collected, namely the EORTC-QLQ-C30, SF-36v2 and EQ-5D-5L instruments. Detailed results from the EORTC-QLQ-C30 and SF-36v2 are presented in Appendix M. In summary, DBCd was associated with substantial benefits to patients based on all the above mentioned measures. During Cycles 1–6 of treatment, DBCd was associated with no decrement in overall HRQoL, fatigue, and mental health, as measured by EORTC-QLQ-C30 Global Health Status, EORTCQLQ-C30 LS mean Fatigue and SF-36v2 Mental Component Summary (MCS) scores. In contrast, BCd was associated with worsening HRQoL during Cycles 1–6. In the context of evidence highlighting the overall HRQoL impairment experienced by AL amyloidosis patients

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

relative to the general population, as well as the high proportion of patients experiencing negative impacts on their mental health, the improvements to HRQoL associated with DBCd are expected to provide considerable value for many patients.[77, 79]

EQ-5D-5L

Interim analysis (IA1): Data cut-off 14[th] February 2020

EQ-5D-5L scores worsened in the BCd group during Cycles 1–6, whereas they remained relatively stable in the DBCd group (Figure 10). At Week 16 (Cycle 4), there was no change in LS **** ****** ***** mean EQ-5D-5L utility scores in the DBCd group ( points; 95% CI: , ), whereas ****** scores decreased (i.e., worsened) significantly in the BCd group (-0.056 points; 95% CI: , ******; unadjusted ******** vs DBCd). After Cycles 1–6, mean EQ-5D-5L utility scores continued to improve in the DBCd group throughout subsequent daratumumab SC monotherapy (Figure 10).

Figure 10: Mean EQ-5D-5L utility scores over time; ITT analysis set (14[th] February 2020 data cut-off)

==> picture [430 x 280] intentionally omitted <==

Abbreviations: C: cycle; CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); Dara SC: daratumumab subcutaneous; D: day; SE: standard error. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

There was also a significant difference in the mean change from baseline at Week 16 for the VAS score, again in favour of DBCd treatment (VAS LS mean change: DBCd: ***** [95% CI: ***** **** ***** ***** ***** ******** , ]; BCd: [95% CI: , ]; ).

These results support that DBCd is a well-tolerated regimen for AL amyloidosis. The addition of daratumumab SC to BCd was not found to have a detrimental effect on HRQoL, with this tolerability being maintained during Cycles 1–6 whilst patients received daratumumab in combination with BCd. Furthermore, improvements to HRQoL are observed once patients then begin the daratumumab SC monotherapy phase of treatment following Cycle 6.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

The EQ-5D-5L utility scores results demonstrate that patients treated with DBCd experience improvements across important aspects of their quality of life. With the questionnaire measuring patients’ level of mobility, general ability to look after themselves and conduct their daily activities, as well as their experience of pain, discomfort, anxiety and depression, the overall improvement in EQ-5D-5L utility scores suggests that treatment with DBCd is associated with improvements across at least some of these important aspects of quality of life. In the context of evidence highlighting the level of impact AL amyloidosis can have on patient quality of life,[77, 79] the improvements to HRQoL observed with DBCd treatment are expected to offer substantial benefit to patients.

As previously described, additional HRQoL outcomes were also assessed at the IA1 analysis, including the EORTC QLQ-C30 and SF-36v2 instruments, with the results presented in Appendix M.

Subgroup analysis

==> picture [34 x 11] intentionally omitted <==

Results from subgroup analyses of the primary efficacy outcome, CHR, from both the prespecified interim analysis and 12-month landmark analyses are presented below.

Interim analysis (IA1): Data cut-off 14[th] February 2020

Achievement of CHR was consistent across all pre-specified, clinically relevant subgroups, including baseline characteristics such as age, sex, race, renal function, hepatic function, and body weight, with greater rates of CHR achieved in the DBCd group compared to the BCd group for all analyses (Figure 11 and Figure 12). CHR rates achieved across body weight categories (≤65 kg, >65 to 85 kg, >85 kg) achieved in the DBCd arm were consistent with the overall population, whereas a lower CHR rate was observed in patients who were ≤65 kg in the BCd arm compared to the overall population. Of the 5 patients that were >120 kg, 2 of the 3 patients in the DBCd arms achieved CHR, while none of the 2 patients in the BCd arm achieved CHR.

When stratified by the severity of cardiac involvement (a key prognostic factor) at baseline, patients in the DBCd group had similar rates of CHR across each cardiac stage (Stage I: 44.7%; Stage II: 53.9%; Stage IIIa/IIIb: 58.3%; Figure 11). In contrast, in the BCd group, the proportion of patients achieving CHR declined as cardiac involvement worsened, ranging from 27.9% at Stage I to just 10.0% at Stage IIIa/IIIb. The CHR rate was similarly high for patients both with and without translocation t(11;14) in the DBCd group (present: 54.9%; absent: *****), whereas the CHR rate was lower among patients with this translocation in the BCd group (present: 12.7%; absent: *****; Figure 12).[95, 98]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Figure 11: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 1 of 2) (14[th] February 2020 data cut-off)

==> picture [376 x 412] intentionally omitted <==

Abbreviations: CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); CI: confidence interval; Dara SC: daratumumab subcutaneous; EVT: event; IRC: independent review committee; ITT: intention-to-treat. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 12: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 2 of 2) (14[th] February 2020 data cut-off)

==> picture [378 x 384] intentionally omitted <==

Abbreviations: CyBorD: cyclophosphamide, bortezomib, dexamethasone (otherwise referred to as BCd); CI: confidence interval; Dara SC; daratumumab subcutaneous; ECOG: eastern cooperative oncology score; EVT: event; FISH: fluorescence in situ hybridisation; IRC: independent review committee; ITT: intention-to-treat. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off).[98]

12-month landmark analysis: Data cut-off 13[th] November 2021

As per the interim analysis, DBCd continued to be associated with a greater achievement of CHR than BCd across all subgroup analyses in the 12-month landmark analysis (Figure 13Figure 14) including patients with Mayo Cardiac Stage III disease or t(11;14) translocation. When stratified by the severity of cardiac involvement at baseline, patients in the DBCd group had similar rates of CHR across each cardiac stage (Stage I: 51.1%; Stage II: 60.5%; Stage IIIa/IIIb: 62.5%; Figure 13). In contrast, similar to the results from the interim analysis, achievement of CHR declined in the BCd group with cardiac involvement worsening, ranging from 30.2% at Stage I to just 10.0% at Stage IIIa/IIIb. Finally, patients in the DBCd group continued to show similarly high CHR rates regardless of t(11;14) translocation (present: 58.8%; absent: *****), whereas patients ***** in the BCd group with this translocation had lower CHR rates (present: 12.7%; absent: ).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 13: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 1 of 2) (13[th] November 2020 data cut-off)

==> picture [385 x 156] intentionally omitted <==

==> picture [385 x 156] intentionally omitted <==

Cardiac stage IIIa/IIIb includes both IIIa subjects and subjects that are IIIa at randomisation and progressed to IIIb at Cycle 1 Day 1. Abbreviations: CHR: complete haematologic response; CI: confidence interval; D-VCd: daratumumab in combination with bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as DBCd); IRC: independent review committee; VCd: bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as BCd). Source: Kastritis et al., (2021).[100]

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Figure 14: Forest plot of subgroup analysis of confirmed CHR based on IRC assessment; ITT analysis set (part 2 of 2) (13[th] November 2020 data cut-off)

==> picture [385 x 137] intentionally omitted <==

==> picture [385 x 136] intentionally omitted <==

Baseline renal stage is defined for subjects with baseline renal involvement. Abbreviations: CHR: complete haematologic response; CI: confidence interval; D-VCd: daratumumab in combination with bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as DBCd); ECOG: ECOG: eastern cooperative oncology score; FISH: fluorescence in situ hybridisation; IRC: independent review committee; VCd: bortezomib, cyclophosphamide and dexamethasone (otherwise referred to as BCd). Source: Kastritis et al., (2021).[100]

==> picture [34 x 12] intentionally omitted <==

Meta-analysis

A clinical SLR conducted (Section B.2.1) identified ANDROMEDA as the only trial analysing the efficacy of DBCd in newly diagnosed AL amyloidosis. It was therefore not necessary to conduct a meta-analysis of multiple trials for DBCd in AL amyloidosis.

==> picture [34 x 12] intentionally omitted <==

Indirect and mixed treatment comparisons

UK clinical experts confirmed that most patients (90-95%) with AL amyloidosis are treated with BCd on the NHS.[88] Further evidence that BCd represents the mainstay of treatment for AL amyloidosis in the UK comes from a retrospective observational analysis conducted by the EMN, where 75% of AL amyloidosis patients in ten countries in Europe were found to receive bortezomib-based regimens as first-line therapy. In line with this evidence, BCd thus represents the sole relevant comparator for this submission.[17]

Given that direct evidence for DBCd compared to BCd is available from the high-quality, RCT ANDROMEDA, it was not necessary to conduct an indirect comparison comparing the efficacy and safety of DBCd with that of other treatments.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

==> picture [42 x 12] intentionally omitted <==

Adverse reactions

Safety results summary  Nearly all patients in both the DBCd and BCd treatment groups reported at least one TEAE (DBCd: 97.9%; BCd: 98.4%) o The most frequently reported TEAEs (≥25%) included diarrhoea, peripheral oedema, constipation, peripheral sensory neuropathy, nausea, fatigue, upper respiratory tract infection, and insomnia  A similar proportion of patients in the DBCd and BCd treatment groups experienced Grade 3 or 4 TEAEs (DBCd: 59%; BCd: 57%) o The most commonly reported (≥5% in either group) included lymphopenia, pneumonia, diarrhoea, neutropenia, syncope, cardiac failure, anaemia, peripheral oedema and hypokalaemia  The proportion of patients reporting serious TEAEs in each treatment group was broadly similar, with a slightly higher proportion reported in the DBCd arm (DBCd: 43.0%; BCd: 36.2%) o The most commonly reported serious TEAEs included pneumonia and cardiac failure/cardiac congestive failure combined  A similarly low proportion of patients in both treatment groups reported AEs that led to discontinuation of study treatment (DBCd: 4.1%; BCd: 4.3%), and the incidence of Grade 3 or 4 AEs that led to discontinuation of treatment was also similarly low across both groups (DBCd: 3.1%; BCd: 2.7%)  The incidence of all grade infusion-related reactions (IRRs) in the DBCd treatment group was 7.4%, which is consistent with that observed in other daratumumab SC studies; all grade injection site reactions were reported in 10.9% of patients treated with daratumumab  At the time of the pre-specified interim analysis, 27 patients (14.0%) had died in the DBCd treatment arm and ** patients (*****) had died in the BCd treatment arm  Overall, DBCd was found to be well-tolerated, with management TEAEs which did not lead to an increase in patient discontinuation and a safety profile consistent with the established safety profiles of daratumumab SC and BCd. Importantly, no new safety concerns were identified.

==> picture [44 x 11] intentionally omitted <==

Treatment duration and dosage

Duration of exposure

A summary of exposure to study treatment in both treatment arms is presented in Table 30. A total of *** patients received at least 1 administration of study treatment. The median number of cycles was 11 months in the DBCd arm and six months in the BCd arm (patients were permitted no more than 6 cycles in the BCd arm). A similar percentage of patients in both treatment arms received study treatment during the first 2 cycles, however from Cycle 3 onwards, more patients in the BCd arm discontinued study treatment compared with subjects in the DBCd arm. In the DBCd arm, ***** of patients completed 6 cycles of treatment compared with ***** of patients in the BCd arm.

With daratumumab SC treatment continuing beyond the initial 6 cycles of BCd, the treatment duration was expected to be longer in the DBCd arm. The median duration of study treatment was 9.6 months for the DBCd arm and 5.3 months for the BCd arm. Among patients in the DBCd arm, ***** received more than 6 cycles of therapy.

Treatment modifications

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

As specified in the trial protocol, dose reductions or escalations were not permitted for daratumumab, with daratumumab-related toxicities being managed by dose delays or dose skipped.

In compliance with the protocol, there were no dose reductions for daratumumab SC, though there were dose delays in **** of patients, as well as doses skipped in ***** of patients. The proportion of patients with dose delays of cyclophosphamide, bortezomib, and dexamethasone was low in both treatment arms (cyclophosphamide: **** vs ****; bortezomib: **** vs ****; dexamethasone: **** vs ****, respectively).

More subjects in the DBCd arm had skipped doses of cyclophosphamide, bortezomib, and dexamethasone compared with the BCd arm (cyclophosphamide: ***** vs *****; bortezomib: ***** vs *****; dexamethasone: ***** vs *****). The proportion of subjects with dose reductions of cyclophosphamide, bortezomib, and dexamethasone was generally similar between arms (cyclophosphamide: ***** vs *****; bortezomib: ***** and *****; dexamethasone: ***** vs *****, respectively).

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 30: Summary of exposure to study treatment, safety analysis set (14[th] February 2020 data cut-off)

Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; SD: standard deviation. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

==> picture [46 x 11] intentionally omitted <==

Adverse events

Safety results from ANDROMEDA presented in this submission are primarily from the prespecified interim analysis (data cut-off 14[th] February 2020), on the basis that adverse events related to study treatment typically occur close to the beginning of treatment. As such, a longer duration of follow-up is not expected to present any further safety signals.

A number of safety outcomes were also assessed at the 12-month landmark analysis (data cutoff 13[th] November 2020), ahead of presentation of updated safety results from ANDROMEDA at a conference. For example, results for the most commonly reported (≥5%) Grade 3 or 4 TEAEs from this 12-month landmark analysis are presented in the submission, highlighting the change in incidence of reported adverse events once patients complete Cycle 6 and are receiving daratumumab SC monotherapy.

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

For both the pre-specified interim and 12-month landmark analyses, results are presented for the safety population.

Interim analysis (IA1): Data cut-off 14[th] February 2020

Nearly all patients in both treatment arms experienced at least one treatment-emergent adverse event (TEAE) (97.9% in the DBCd arm vs 98.4% in the BCd arm; Table 31). TEAEs occurring with an incidence of greater than 25% were generally balanced between treatment arms, with the exception of peripheral sensory neuropathy and upper respiratory tract infection which occurred at a higher incidence in the DBCd arm. TEAEs leading to discontinuation were also ******** between the treatment arms. Serious TEAEs and Grade 5 TEAEs were *********** ****** in the DBCd arm, reflecting the longer treatment exposure in the DBCd arm and longer TEAE reporting period for patients treated with DBCd.

Company evidence submission template for ID3748 © Janssen-Cilag (2021). All rights reserved

Table 31: Overall summary of treatment-emergent adverse events; safety analysis set (14[th] February 2020 data cut-off)

Toxicity grade is defined according to the NCI CTCAE (National Cancer Institute Common Terminology Criteria for Adverse Events) Version 4.03.[a] TEAEs related to at least 1 of the 4 components of study treatment: cyclophosphamide, bortezomib, dexamethasone and daratumumab.[b] TEAEs leading to discontinuation of all study treatment due to an adverse event on the end of treatment CRF page.[c] Site reporting error: site reported at least 1 AE as related to daratumumab in error, for 1 subject randomised to BCd arm. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; CRF: case report form; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; SD: standard deviation; TEAE: treatment-emergent adverse event. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved Page 85 of 165

Table 32 summarises the most commonly reported (>10%) TEAEs by preferred term in the DBCd and BCd treatment arms. TEAEs in which differences in incidence of more than 5% between treatment arms were observed included diarrhoea, constipation, peripheral sensory neuropathy, upper respiratory tract infection, dyspnoea, thrombocytopenia, cough, asthenia, back pain and arthralgia. While these were all reported with greater incidence in the DBCd arm than in the BCd arm, it should be considered that the median treatment duration for DBCd was substantially longer (9.6 months) than the median treatment duration for BCd (5.3 months).

Table 32: Most commonly reported (>10% in either arm) treatment-emergent adverse events by preferred term; safety analysis set (14[th] February 2020 data cut-off)

Adverse events are reported using MedRA Version 22.1. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; TEAE: treatment-related adverse event. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

Table 33 presents a summary of the most common (≥5% in either treatment arm) Grade 3 or 4 TEAEs experienced by patients in the DBCd and BCd treatment arms. Grade 3 or 4 TEAEs were

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

reported in 58.5% of patients in the DBCd arm and 57.4% of patients in the BCd arm and were well-balanced between arms.

Table 33: Most commonly reported (>5%) toxicity Grade 3 or 4 treatment-emergent adverse events by system organ class and preferred term; safety analysis set (14[th] February 2020 data cut-off)

Patients are counted only once for any given event, regardless of the number of times they actually experienced the event. Adverse events are reported using MedRA Version 22.1. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; TEAE: treatment-related adverse event. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020)[95]

A summary of the most common treatment-emergent serious adverse events is presented in Table 34. In the DBCd treatment arm, 43% of patients reported ≥1 treatment-emergent serious adverse event (SAE) compared with 36.2% of patients in the BCd arm. The most commonly (≥5% in either treatment arm) reported treatment-emergent SAEs included pneumonia (DBCd 7.3%; BCd: 4.8%) and cardiac failure/cardiac failure congestive combined (DBCd: 6.7% [13/193]; BCd: 5.3% [10/188]). A difference in the incidence of >2% between treatment arms was observed for the following treatment-emergent SAEs: pneumonia (DBCd: 7.3%; BCd: 4.8%) and sepsis (3.1% and 0%, respectively) and cardiac arrest (3.6% and 1.6%, respectively; Table 33). Fluid overload was reported with ≥2% higher incidence in the BCd arm (DBCd: 0.5%; BCd: 2.7%).

Table 34: Most common (at least 2% in either arm) treatment-emergent serious adverse events by system organ class and preferred term; safety analysis set (14[th] February 2020 data cut-off)

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Patients are counted only once for any given event, regardless of the number of times they actually experienced the event. Adverse events are coded using MedRA Version 22.1. Abbreviations: BCd: bortezomib, cyclophosphamide and dexamethasone; DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone; TEAE: treatment-related adverse event. Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020);[95] ANDROMEDA Study Results (clinicaltrials.gov).[103]

With use of the SC formulation of daratumumab in ANDROMEDA, it was necessary to explore the extent of IRRs relating to DBCd treatment. A summary of the proportion of patients in the DBCd treatment group that experienced treatment-emergent infusion-related reactions is presented in Table 35.

Of 193 patients who received DBCd, 7.3% experienced an IRR. IRRs were Grade 1 or 2 (manageable) and did not lead to treatment discontinuation. The incidence, preferred terms, severity and onset of IRRs were consistent with those previously reported for daratumumab SC.

Table 35: Number of patients with treatment-emergent infusion-related reactions by system organ class, preferred term and maximum toxicity grade; safety analysis set (14[th] February 2020 data cut-off)

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

Adverse events are reported using MedRA Version 22.1. Toxicity grade is defined according to the NCI CTCAE (National Cancer Institute Common Terminology Criteria for Adverse Events) Version 4.0. Abbreviations: DBCd: daratumumab SC in combination with bortezomib, cyclophosphamide and dexamethasone.

Source: Janssen ANDROMEDA CSR (14[th] February 2020 data cut-off);[98] Kastritis et al., (2020).[95]

Deaths

27 patients (14.0%) died in the DBCd group, whilst ** patients (*****) died in the BCd group. In addition, one patient randomised to BCd died prior to receiving any treatment. Overall, deaths were primarily due to AL amyloidosis-related cardiomyopathy, reported either as TEAEs or disease progression, in both treatment arms. Cardiac disorders were the primary cause of death in both treatment groups, most commonly from cardiac arrest or failure (DBCd: **** and ****, respectively; BCd: **** and ). Compared with the BCd group, more patients in the DBCd group died from TEAEs (* and , respectively), with fewer patients dying from disease progression ( and ****). Nearly all patients who died from TEAEs had cardiac involvement at baseline (DBCd: ***** patients; BCd: ***** patients); relatively few deaths from cardiac events or other causes were considered related to study treatment (DBCd: overall: ****; cardiac: ****; BCd:

Company evidence submission template for ID3748

© Janssen-Cilag (2021). All rights reserved

overall: ****; cardiac: ****). No patients with Mayo Cardiac Stage I disease at baseline died due to a TEAE during follow-up.

12-month landmark analysis: Data cut-off 13[th] November 2020

In the 12-month landmark analysis (median follow-up 20.3 months), both DBCd and BCd remained well-tolerated, with no new safety concerns identified. As with the interim analysis, safety results should continue to be interpreted in the context of different treatment durations in the DBCd and BCd patient groups. The DBCd group continued to receive daratumumab SC monotherapy from Cycle 7 onwards and the median treatment duration was 18.5 months in this group, 3.5 times longer than the 5.3 months in the BCd group.

A summary of commonly reported (in ≥5% of patients) Grade 3 or 4 TEAEs is presented in Table 36. Grade 3 or 4 TEAEs were reported by ***** of patients in the DBCd group (56.0% for Cycles 1–6) and 57.4% of patients in the BCd group. From Cycle 7 onwards, when patients in the DBCd treatment arm were receiving daratumumab SC monotherapy, no Grade 3 or 4 TEAEs were reported in ≥5% of patients.

Table 36: Most commonly reported (≥5% in either arm) toxicity Grade 3 or 4 treatmentemergent adverse events by system organ class and preferred term; safety analysis set (13[th] November 2020 data cut-off)