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

Single Technology Appraisal

Tafasitamab with lenalidomide for treating relapsed or refractory diffuse large B-cell lymphoma [ID3795]

Committee Papers

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

NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE

SINGLE TECHNOLOGY APPRAISAL

Tafasitamab with lenalidomide for treating relapsed or refractory diffuse large B-cell lymphoma [ID3795]

Contents:

The following documents are made available to consultees and commentators:

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

Pre-technical engagement documents

1. Company submission from Incyte

2. Clarification questions and company responses

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

• a. Lymphoma Action

• b. National Cancer Research Institute-Association of Cancer PhysiciansRoyal College of Physicians-Royal College of Radiologists

4. Evidence Review Group report prepared by Kleijnen Systematic Reviews

5. Evidence Review Group report – factual accuracy check

Post-technical engagement documents

6. Technical engagement response from company

7. Technical engagement responses and statements from experts:

• a. Professor Andrew Davies – clinical expert, nominated by Incyte

• b. Dr Kate Cwynarski – clinical expert, nominated by Incyte and National Cancer Research Institute-Association of Cancer Physicians-Royal College of Physicians-Royal College of Radiologists

8. Technical engagement responses from consultees and commentators:

• a. National Cancer Research Institute-Association of Cancer PhysiciansRoyal College of Physicians-Royal College of Radiologists

9. Evidence Review Group critique of company response to technical engagement prepared by Kleijnen Systematic Reviews

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

Tafasitamab with lenalidomide for treating relapsed or refractory diffuse large B-cell lymphoma [ID3795]

Document B

Company evidence submission

May 2022

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Contents

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Abbreviations

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Tables and figures

Table 1. The decision problem .............................................................................................. 15 Table 2. Technology being appraised ................................................................................... 18 Table 3. Treatment response in patients with R/R DLBCL by line of treatment (Christie National Health Service Foundation Trust Database, 2011 to 2017) .................................... 25 Table 4. Clinical-effectiveness evidence—L-MIND (MOR208C203)..................................... 30 Table 5. Clinical-effectiveness evidence—MOR208C201 .................................................... 31 Table 6. L-MIND methodology .............................................................................................. 32 Table 7. MOR208C201 methodology ................................................................................... 36 Table 8. Quality assessment for L-MIND (MOR208C203) and MOR208C201 ..................... 40 Table 9. L-MIND study: selected demographics and baseline characteristics ...................... 42 Table 10. MOR208C201 study: selected demographics and baseline characteristics–DLBCL cohort (FAS) .......................................................................................................................... 45 Table 11. Best ORR (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) 46 Table 12. Primary efficacy analysis: MOR208C201 (ITT population) ................................... 54 Table 13. RE-MIND study: overview of efficacy outcomes–MAS25 ..................................... 56 Table 14. Inclusion and exclusion criteria for the RE-MIND2 study ...................................... 58 Table 15. Baseline covariates used in the ePS for RE-MIND2 ............................................. 59 Table 16. Studies identified for the MAIC study by the SLR and clinician interviews ........... 66 Table 17. TEAEs (SAS) ........................................................................................................ 76 Table 18. TEAEs (SAS) ........................................................................................................ 80 Table 19. Clinical development programme for tafasitamab ................................................. 82 Table 20. L-MIND extended follow-up analysis for PFS, DOR and OS by prior lines of therapy (data cut-off 30 October 2020; ≥35 months of follow-up){Duell, 2021 #232} ........... 86 Table 21. End-of-life criteria .................................................................................................. 88 Table 22. Overview of cost-effectiveness analysis studies ................................................... 90 Table 23. Key features of the economic analysis ................................................................. 96 Table 24. Base case modelling approaches for OS and PFS ............................................. 101 Table 25. Summary of base-case inputs ............................................................................ 104 Table 26. Key Assumptions ................................................................................................ 106 Table 27. HRQoL and utility studies in R/R DLBCL identified in the SLR and previous NICE appraisals for R/R DLBCL ................................................................................................... 107 Table 28. Summary of utility values for cost-effectiveness analysis ................................... 109 Table 29. AE Disutilities ...................................................................................................... 110 Table 30. Induction Drug Costs .......................................................................................... 114 Table 31. Maintenance Drug Costs .................................................................................... 115 Table 32. Induction Treatment Schedule ............................................................................ 116 Table 33. Maintenance Treatment Schedule ...................................................................... 117 Table 34. Administration Costs ........................................................................................... 117 Table 35. Co-medication Drug Dosing and Cost Calculation .............................................. 118 Table 36. Administration Dosing for Co-medications .......................................................... 119 Table 37. Co-medication Costs ........................................................................................... 120 Table 38. Subsequent Treatment Distributions ................................................................... 120 Table 39. Subsequent Treatment Drug Costs .................................................................... 123 Table 40. Total Subsequent Treatment Costs .................................................................... 127 Table 41. Unit Costs for Monitoring Tests ........................................................................... 127 Table 42. Monitoring Tests: Frequency of Use per Cycle (PFS patients without prolonged PFS) .................................................................................................................................... 129

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Table 43. One-off Monitoring Cost ...................................................................................... 130 Table 44. Monitoring Cost per Cycle (PFS patients without Prolonged PFS) ..................... 130 Table 45. Monitoring Costs: Frequency of Use per Model Cycle (by year of Prolonged PFS status) ................................................................................................................................. 131 Table 46. Monitoring Cost per Cycle: Prolonged PFS patients ........................................... 131 Table 47. Disease Management Resource Unit Cost ......................................................... 131 Table 48. Disease Management: Frequency of Use per Model Cycle (PFS without prolonged PFS) .................................................................................................................................... 133 Table 49. Disease Management Cost per Cycle for PFS patients without prolonged PFS 133 Table 50. Disease Management: Frequency of Use Per Cycle (Prolonged PFS) .............. 134 Table 51. Disease Management Cost per Cycle: Prolonged PFS ...................................... 134 Table 52. Disease Management: Frequency of Use (Progressed) ..................................... 135 Table 53. Disease Management Cost per Cycle: Post Progression ................................... 135 Table 54. One-off Costs ...................................................................................................... 136 Table 55. Cumulative Probability of AEs during the Treatment Period ............................... 137 Table 56. Cost of Managing AEs per Event ........................................................................ 137 Table 57. AE Management Costs per Treatment ............................................................... 138 Table 58. Base-case results ............................................................................................... 138 Table 59: Base case results – full incremental analysis ..................................................... 139 Table 60. Mean PSA results ............................................................................................... 140 Table 61. Scenario analysis results .................................................................................... 146

Figure 1. NICE-recommended treatment pathway for R/R DLBCL – updated to reflect current UK clinical practice ................................................................................................................ 23 Figure 2. Current NICE recommendations for patients with R/R DLBCL who are not eligible for transplant ......................................................................................................................... 24 Figure 3. Tafasitamab mechanism of action ......................................................................... 28 Figure 4. Proposed place for tafasitamab in the pathway of care for patients with R/R DLBCL who are transplant ineligible – updated to reflect current UK clinical practice ...................... 29 Figure 5. L-MIND study: patient disposition (all patients enrolled) ........................................ 41 Figure 6. KM plot of DoR (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) .............................................................................................................................. 47 Figure 7. KM plot of DoR by best objective response (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) ...................................................................................... 48 Figure 8. KM plot of PFS (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) .............................................................................................................................. 49 Figure 9. KM plot of OS (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) .............................................................................................................................. 51 Figure 10. KM plot of OS by best objective response (updated analysis data cut-off 30 October 2020; FAS; IRC assessed) ...................................................................................... 52 Figure 11. KM plot for OS: BR (a) and R-GemOx (b) ........................................................... 61 Figure 12. Forest plot of OS HRs with 95% CIs using Cox proportional hazard model for different analysis sets ........................................................................................................... 62 Figure 13. KM estimates for OS for TAFA+LEN observed (green) and adjusted (blue) compared with reported OS estimates for comparators (red) ............................................... 68 Figure 14. KM estimates for PFS for TAFA+LEN observed (green) and weighted (blue) compared with reported PFS-IRC estimates for comparators (red) ...................................... 69

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Figure 15. KM estimates for DoR for TAFA+LEN observed (green) and weighted (blue) compared with reported DoR estimates for comparators (red) ............................................. 71 Figure 16. Depth of IRC responses for TAFA+LEN observed (green) and weighted (blue) compared with those reported for comparators (red) ............................................................ 73 Figure 17. Model Diagram .................................................................................................... 93 Figure 18. Example survival partition approach .................................................................... 94 Figure 19. Base case OS extrapolations ............................................................................ 102 Figure 20. Base case PFS extrapolations ........................................................................... 103 Figure 21. General population and age/sex adjusted health state utility curves ................. 111 Figure 22. PSA cost-effectiveness plane for TAFA+LEN vs. pola-BR ................................ 141 Figure 23. PSA cost-effectiveness plane for TAFA+LEN vs. BR ........................................ 141 Figure 24. PSA cost-effectiveness plane for TAFA+LEN vs. R-GemOx ............................. 142 Figure 25. CEAC ................................................................................................................. 142 Figure 26. Tornado diagram of ICER results for TAFA+LEN vs. pola-BR .......................... 143 Figure 27. Tornado diagram of ICER results for TAFA+LEN vs. BR .................................. 144 Figure 28. Tornado diagram of ICER results for TAFA+LEN vs. R-GemOx ....................... 145

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

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

B.1.1. Decision problem

The submission covers the technology’s full marketing authorisation for this indication.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Table 1. The decision problem

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

Abbreviations: AE = adverse event; ASCT = autologous stem cell transplant; BR = rituximab in combination with bendamustine; CR = complete response; DLBCL = diffuse ‐ large B cell lymphoma; NHS = National Health Service; NICE = National Institute for Health and Care Excellence; OS = overall survival; PFS = progression-free survival; polaBR = polatuzumab vedotin with bendamustine and rituximab; PR = partial response; R-DECC = rituximab, dexamethasone, etoposide, chlorambucil, lomustine; PSS = Personal Social Services; R-GemOx = rituximab in combination with gemcitabine and oxaliplatin; R-Gem = rituximab in combination with gemcitabine; R-P-MitCEBO = rituximab, prednisolone, mitoxantrone cyclophosphamide, etoposide bleomycin, vincristine; SAE = serious adverse event; TTD = time to treatment discontinuation or death; UK = United Kingdom

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL [ID3795]

©Incyte(2022). All rights reserved

B.1.2. Description of the technology being appraised

A summary of tafasitamab is shown in Table 2, and the draft summary of product characteristics is included in Appendix C.

Table 2. Technology being appraised

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Abbreviations: ADCC = antibody-directed cellular cytotoxicity; ASCT = autologous stem cell transplant; CE = ‐ cost-effectiveness; DLBCL = diffuse large B cell lymphoma; EMA = European Medicines Agency; Fc = fragment crystallisable; NK = natural killer; SmPC = summary of product characteristics; UK = United Kingdom

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

B.1.3.1. Disease overview

Non-Hodgkin’s lymphoma (NHL) is a heterogeneous group of haematological malignancies that originate in the lymphocyte cells of the immune system.(11) Approximately 90% of NHL originates from B cells (B-cell lymphoma) and the remaining cases of NHL originate from T cells or natural killer (NK) cells. There are at least 30 subtypes of mature B-cell NHL malignancies, which are classified into high- and low-grade NHL subtypes.(11) The high-grade subtypes have a worse prognosis than the low-grade forms. Diffuse large B-cell lymphoma (DLBCL) is a high-grade subtype of B-cell NHL.(11)

DLBCL is classified as a rare disease, and represents approximately 40% of all newly diagnosed NHL cases.(11-13) DLBCL is composed of large neoplastic B lymphoid cells expressing pan B-cell antigens, including CD19 and CD20.(14) While there is no single cytogenetic change that is typical or diagnostic of DLBCL, genetic abnormalities are common.(14) As a result, treatment is focused on B-cell antigen expression (Section B.1.3.5. ).

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

B.1.3.2. Epidemiology

DLBCL affects approximately 2.5 in 10,000 people in the European Union (EU).(15) In the United Kingdom (UK), the Office of National Statistics (ONS) suggests that there will be approximately 4,826 new cases of DLBCL each year. Patients with newly diagnosed DLBCL are generally older (median age of 66 years) and there is a slightly higher incidence of DLBCL in men.(11, 16, 17)

B.1.3.3. Prognosis

Although DLBCL is aggressive if left untreated, patients display high response rates to chemotherapy in the first line (1L), ranging from 88% to 91% depending on the classification system used. In the UK, the five-year survival rate for patients with 1L DLBCL therapy is approximately 61%.(18)

Despite good initial response rates, between 10% and 20% of patients with DLBCL are refractory to standard 1L chemotherapy,(19-22) and another 30% of patients will ultimately relapse.(23, 24) There has been limited improvement in the survival of adults with DLBCL at subsequent lines of therapy.

In patients with relapsing DLBCL, less than half of patients will survive the 12 months following diagnosis (41%; median survival, 10 months).(25) Age is an important prognostic indicator in patients with DLBCL who relapse—patients aged ≥65 years have a worse prognosis than those younger than 65.(25)

Prognosis is worse for patients who are refractory to 1L therapy. Median overall survival was 6.3 months, with only 22% of patients alive at two years, in a large pooled retrospective analysis of patients with refractory DLBCL (SCHOLAR-1 study).(26)

Patients with relapsed/refractory (R/R) DLBCL have a worse prognosis and a greater symptomatic burden than patients with newly diagnosed DLBCL due to the progressive nature of the disease and the cumulative adverse effects of intensive treatment.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

B.1.3.4. Disease burden

Patient burden

Patients with DLBCL typically present with a rapidly enlarging lymphadenopathy, most commonly a nodal enlargement in the neck or abdomen, and systemic symptoms that require immediate treatment.(27) Systemic "B" symptoms (i.e., fever, weight loss, drenching night sweats, fatigue and pruritus) are observed in approximately 30% of patients.(12, 20)

Approximately 60% of patients will present with advanced-stage DLBCL (Ann Arbor stage III or IV disease). In approximately 40% of cases, the disease arises in extranodal medullary tissues.(28)

While data on the impact of DLBCL on patients’ quality of life (QoL) are limited, it is well established that patients with high-grade NHL demonstrate a lower QoL compared with patients with low-grade NHL, including physical, social/family, emotional factors and functional well-being.(29) Patients with high-grade NHL also demonstrate higher levels of anxiety than patients with low-grade NHL.(29) The negative impact of high-grade NHL on patient QoL has been attributed to(30):

• Uncertainties around disease prognosis

• Side effects of treatment

• Fear of relapse

Patients who achieve a complete response (CR) after 1L treatment have demonstrated significant improvements in QoL compared with patients not achieving a CR.(31) Patients who are relapsed or refractory to first line treatment experience worse health-related QoL (HRQoL) due to the poorer prognosis of their condition and the need for additional, often more intensive subsequent treatment.(31). Achieving a CR even in later lines of treatment is therefore a key treatment goal in patients with R/R DLBCL.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Healthcare burden

DLBCL is the most costly lymphoma to treat in Europe, when compared with Hodgkin’s lymphoma and follicular lymphoma. This is mainly driven by inpatient hospital stays, medication, and productivity loss.(32) DLBCL treatment across all lines of therapy is complex, involving multiple sites of care and treatment types.(33)

In a prevalence-based estimate of costs in the UK, the total cost associated with treating new patients with DLBCL over a one-year period was approximately £88 to £92 million.(34) However, limited cost studies have been completed for treatments used in later lines.

B.1.3.5. Clinical pathway of care

The treatment pathway for patients with DLBCL, including R/R DLBCL, is provided by National Institute for Health and Care Excellence (NICE) 2016 guidance NG52, the British Society for Haematology (BSH), the European Society for Medical Oncology (ESMO) and the National Comprehensive Cancer Network (NCCN).(1, 5, 6, 24, 35, 36) Subsequent to the publication of these guidelines, new treatments have become available which have been included in the treatment pathway.(1, 5, 6)

An overview of the treatment pathway is shown in Figure 1.

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

Figure 1. NICE-recommended treatment pathway for R/R DLBCL – updated to reflect current UK clinical practice

==> picture [512 x 119] intentionally omitted <==

==> picture [512 x 119] intentionally omitted <==

Sources: NICE guidance NG52;(36) NICE technology appraisal (TA)649;(5) NICE TA567;(6) NICE TA559;(1) NICE TA306;(35) Tilly 2015(24)

Abbreviations: ASCT = autologous stem cell transplant; BR = bendamustine with rituximab; CAR-T = chimeric antigen receptor T-cell; R = rituximab; R-CHOP = rituximab in combination with cyclophosphamide, doxorubicin, vincristine, and prednisone; R/R = relapsed/refractory

1L treatment

SoC 1L therapy for DLBCL is chemoimmunotherapy, usually comprising rituximab in combination with cyclophosphamide, doxorubicin, vincristine, and prednisone chemotherapy (R-CHOP).(24, 36)

2L treatment

Patients who relapse or are refractory to 1L treatment have a poor prognosis and few available and effective treatment options.(37, 38) The first step of R/R DLBCL treatment is to assess whether the patient is fit for intensive salvage therapy and potentially for autologous stem cell transplant (ASCT).(38)

Approximately 50% of patients are not transplant eligible, either because they are: 1) chemo-refractory to salvage chemotherapy administered prior to ASCT; 2) they have advanced disease or comorbidities, severe concomitant medical or psychiatric illness, active central nervous system involvement or human immunodeficiency virus (HIV) seropositivity; or 3) they have treatment failure following a prior ASCT.(39, 40)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Not all eligible patients go on to receive a transplant. Retrospective studies have shown that only 25% to 38% of patients who relapsed following rituximab chemotherapy underwent ASCT.(24, 39)

Transplant-ineligible patients

There is no clear SoC for patients with R/R DLBCL who are unable to tolerate intensive therapy or are ineligible for ASCT. As the guidelines were developed prior to the availability of newer targeted therapies, such as polatuzumab vedotin with bendamustine and rituximab (pola-BR) and chimeric antigen receptor T-cell (CART)-cell therapies, the suggested treatment options for patients who relapse and are not eligible for transplant are clinical studies with novel drugs or palliative care. Current NICE recommendations for patients with R/R DLBCL who are ineligible for transplant in the UK are shown in Figure 2. The treatment goal remains the same across the guidelines and guidance’s being improving and prolonging survival.(41)

Figure 2. Current NICE recommendations for patients with R/R DLBCL who are not eligible for transplant

==> picture [446 x 121] intentionally omitted <==

==> picture [446 x 121] intentionally omitted <==

Source: Adapted from NICE pathways: Treating diffuse large B-cell lymphoma(41) Abbreviations: 1L = first line; 2L+ = second line or later; 3L = third line; 3L+ = third line or later; 4L = fourth line; DLBCL - diffuse large B-cell lymphoma; SCT = stem cell transplant; TA = technology appraisal

SCHOLAR-1 is the largest international, retrospective, patient-level, pooled-analysis to evaluate response and survival rates in patients with R/R-DLBCL. These data

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

are particularly important because they represent a large number of patients treated in the modern rituximab era. Patient-level data were collected from medical records for patients with refractory DLBCL.(26)

The study (pooled N=636) revealed a median overall survival (OS) of 6.3 months (95% confidence interval [CI]: 5.9, 7.0 months), with a one-year survival rate of 28% and a two-year OS of 20%. Patients achieved a response rate (RR), CR and partial response (PR) of 26% (95% CI: 21%, 31%), 7% (95% CI: 3%, 15%), and 18% (95% CI: 13%, 23%), respectively.(26) The data show that even with the availability of multiple rituximab-based regimens, outcomes among patients with R/R DLBCL remain dismal—a finding which underlines the high unmet need of this patient population.

A recent systematic review by Vander Velde et al., (2019),(42) identified 19 studies of patients with R/R DLBCL, of which six studies were randomised controlled trials (RCT) and 13 were prospective, observational, single-arm trials.(42) The review reported a median progression-free survival (PFS) range of 2.6 to 17.1 months (n=11 studies) and an OS of 5.0 to 22.2 months (n=11 studies) in patients with R/R DLBCL. It further concluded that there was a paucity of

published RCTs demonstrating comparative efficacy of R/R DLBCL treatments which in turn reflected the lack of proven treatment options in this stage of the pathway.(42)

A UK, single-centre, retrospective analysis of patients with DLBCL who had an R/R event demonstrated an objective response rate (ORR) of 46.1% in the 2L, 27.0% in the 3L, and 9.8% in the fourth line (4L) and later. Overall, patients with R/R DLBCL had a two-year OS of 30.6%.(43) Detailed response rates are shown in Table 3.

Table 3. Treatment response in patients with R/R DLBCL by line of treatment (Christie National Health Service Foundation Trust Database, 2011 to 2017)

Source: Radford et al, 2019(43)

Abbreviations: 2L = second line; 3L = third line; 4L+ = fourth line and later; CI = confidence interval; CR = complete response; OS = overall survival; PR = partial response

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Pola-BR

ESMO and BSH recommendations were developed prior to the availability of pola-BR.(24, 44) In the 2L setting, patients who are transplant ineligible may now receive pola-BR.

Polatuzumab vedotin (pola) is a CD79b-targeted antibody drug conjugate delivering a microtubule inhibitor. CD79b is a signalling component of the B-cell receptor located on most mature B-cell malignancies, including >95% of DLBCL.

Pola-BR was compared with BR in a randomly assigned multicohort of patients (N=80) with transplant-ineligible R/R DLBCL. Patients aged ≥18 years were eligible if they had biopsy-confirmed R/R DLBCL (excluding transformed lymphoma) after ≥1 prior line of therapy, an Eastern Cooperative Oncology Group (ECOG) performance status score of 0 to 2, grade ≤1 peripheral neuropathy, and were considered transplantation ineligible by the treating physician or experienced treatment failure with prior ASCT.(21)

In 40 patients with R/R DLBCL, pola-BR demonstrated an ORR of 45%, a CR rate of 40%, a median PFS of 9.5 months (95% CI: 6.2, 13.9 months), and OS of 12.4 months (9.0 months, not reached). In the pola-BR treatment arm, 33.3% of patients discontinued all treatment due to adverse events (AE), most commonly

thrombocytopaenia and neutropaenia. Peripheral neuropathy (including peripheral motor neuropathy, peripheral sensory neuropathy, decreased vibratory sense, hypaesthesia and paraesthesias) occurred in 43.6% of patients in the pola-BR combination treatment arm (all grades 1 to 2) and resulted in treatment delays in one patient.(21)

Pola-BR has some limitations. The treatment targets the CD20 antigen, which has been shown to undergo a negative transformation (or loss of expression) in up to 60% of patients after treatment with rituximab-containing chemotherapy.(45-48) Therefore, pola-BR may not be appropriate for treatment in this potentially large proportion of patients who experience a loss of CD20 antigen expression after rituximab therapy.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Subsequent lines of treatment

CAR T-cell therapies and pixantrone monotherapy are currently funded by NICE in the 3L setting.

CAR-T therapy may be offered via the Cancer Drugs Fund if the patient is healthy enough to undergo the treatment and has had ≥2 lines of prior systemic therapy.(24, 49)

While pixantrone monotherapy is currently recommended by NICE in the 3L and 4L settings, limited efficacy data in the real world (median OS 3.4) have restricted its use in clinical practice.(50) In addition, interviews with clinical experts in the UK did not consider pixantrone a suitable treatment option in this patient population.{Incyte Corporation, 2020 #316}

B.1.3.6. Tafasitamab and its place in therapy

Patients with R/R DLBCL who are ineligible for transplant or who relapse after transplant have no established SoC. While the treatment aim remains to improve and prolong survival, recent data suggests poor overall survival with currently available treatment options.(43)

Tafasitamab is a novel treatment that has shown efficacy as a single agent in patients with DLBCL (Section B.1.2. ). Tafasitamab was granted orphan designation by the European Medicines Agency (EMA) in 2014; in 2021, orphan designation was maintained by the EMA and granted by the MHRA for tafasitamab for the treatment of DLBCL.(10)

Tafasitamab is a fragment crystallisable (Fc)-enhanced mAb that targets the CD19 antigen expressed on the surface of pre-B and mature B-lymphocytes across different B-cell malignancies, including DLBCL. Upon binding to CD19, tafasitamab mediates B-cell lysis through the engagement of immune effector cells like NK cells, γδ T cells and phagocytes, and direct induction of cell death (apoptosis). The Fc modification results in enhanced antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP; Figure 3).(7)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Preclinical data suggested that tafasitamab acts synergistically with lenalidomide, an immunomodulatory agent that enhances the activity and recruitment of NK cells, and that has been shown to enhance NK cell-mediated antibody-directed cellular cytotoxicity).(51-53) The novel mechanism of action of tafasitamab with lenalidomide is an innovative treatment approach that has been demonstrated to be an effective, well-tolerated, immunomodulatory, chemotherapy-free treatment option for patients with R/R DLBCL who are ineligible for ASCT or who have relapsed after ASCT (Section B.2.6. ).(54)

Figure 3. Tafasitamab mechanism of action

==> picture [417 x 113] intentionally omitted <==

==> picture [417 x 114] intentionally omitted <==

==> picture [417 x 114] intentionally omitted <==

Source: Poe et al., 2012(55)

Tafasitamab is indicated in combination with lenalidomide followed by tafasitamab monotherapy for the treatment of adults with R/R DLBCL who are not eligible for ASCT.(9) Figure 4 shows the proposed placement of tafasitamab + lenalidomide (TAFA+LEN) in the treatment of patients with R/R DLBCL who are not eligible for ASCT.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

The following patients could be considered eligible for TAFA+LEN

• R/R 2L patients who are ineligible for ASCT

• R/R 3L patients (or beyond) who are ineligible for ASCT (including those who

  • relapse following ASCT or receive salvage chemotherapy but fail to respond, and are therefore considered transplant ineligible)

Figure 4. Proposed place for tafasitamab in the pathway of care for patients with R/R DLBCL who are transplant ineligible – updated to reflect current UK clinical practice

==> picture [491 x 86] intentionally omitted <==

==> picture [491 x 86] intentionally omitted <==

==> picture [491 x 86] intentionally omitted <==

Sources: NICE guidance NG52;(36) NICE technology appraisal (TA)649;(5) NICE TA567;(6) NICE TA559;(1) NICE TA306;(35) Tilly 2015(24)

Abbreviations: ASCT = autologous stem cell transplant; BR = bendamustine with rituximab; CAR-T = chimeric antigen receptor T-cell; R = rituximab; R-CHOP = rituximab in combination with cyclophosphamide, doxorubicin, vincristine, and prednisone; R/R = relapsed/refractory

The combination therapy of TAFA+LEN followed by tafasitamab monotherapy is being studied in the pivotal L-MIND study described in Section B.2.9. Data supporting long-term maintenance of response with tafasitamab monotherapy following TAFA+LEN are also presented. Evidence of the clinical effectiveness of the combined therapy is also supported by an indirect comparison with the RE-MIND study, a retrospective chart review of patients with R/R disease treated with lenalidomide monotherapy and the RE-MIND2 retrospective chart review study of

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

patients with R/R disease receiving other treatments that are routinely administered in clinical practice. The comparative efficacy results are presented in Section B.2.9.

B.1.4. Equality considerations

There are no known equality issues relating to the use of tafasitamab in patients with R/R DLBCL who are not eligible for ASCT.

B.2. Clinical effectiveness

B.2.1. Identification and selection of relevant studies

A systematic literature review (SLR) was conducted to identify relevant clinical evidence in R/R DLBCL. Searches were conducted on 9 February 2021 and updated on 29 June 2021. A total of nine reports were identified from 32 unique studies. Full details of the process and methods used to identify and select the clinical evidence relevant to the technology being appraised are included in Appendix D.

B.2.2. List of relevant clinical-effectiveness evidence

B.2.2.1. L-MIND phase II study (TAFA+LEN)

The submission is supported by data on the safety and efficacy of TAFA+LEN from the pivotal, phase II, open-label, single-arm, multicentre L-MIND study (MOR208C203; NCT02399085). Data sources for L-MIND included Salles et al., 2020,(53) the L-MIND clinical study reports (CSR),(56, 57) Salles et al., 2020

European Hematology Association,(58), Duell et al., 2021(54) and Incyte data on file. Table 4 summarises the L-MIND study.

Table 4. Clinical-effectiveness evidence—L-MIND (MOR208C203)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

*Outcomes marked in bold are incorporated into the economic model. Source: Salles et al., 2020(53)

Abbreviations: ASCT = autologous stem cell transplantation; DCR = disease control rate; DLBCL = diffuse large B-cell lymphoma; DoR = duration of response; INV = investigator; IRC = independent radiology/clinical review committee; IV = intravenous; NA = not applicable; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; R/R = relapsed or refractory; TAFA+LEN = tafasitamab + lenalidomide; TTNT = timeto-next treatment; TTP = time to progression

B.2.2.2. MOR208C201 phase IIa study – DLBCL cohort (tafasitamab

monotherapy)

Additional supportive data with tafasitamab monotherapy are provided by the DLBCL cohort of the phase IIa, open-label, multicentre MOR208C201 study in patients with R/R B-cell NHL (Table 5).

Table 5. Clinical-effectiveness evidence—MOR208C201

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

*Outcomes marked in bold are incorporated into the economic model.

Source: Incyte, data on file (MOR208C201 CSR)(59)

Abbreviations: CR = complete response; DLBCL = diffuse large B-cell lymphoma; DoR = duration of response; IRC = independent radiology/clinical review committee; IV = intravenous; NA = not applicable; NHL = nonHodgkin’s lymphoma; ORR = objective response rate; PFS = progression-free survival; PR = partial response; R/R = relapsed or refractory; TAFA+LEN = tafasitamab + lenalidomide; TTP = time to progression

B.2.3. Summary of methodology of the relevant clinicaleffectiveness evidence

A summary of the L-MIND methodology is provided in Table 6.

Table 6. L-MIND methodology

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

*Note: The definition of primary refractory DLBCL was revised (Protocol Amendment 2, Final Version 5.0 [27 Jun 2016]), (less than a PR to 1L therapy or progression within six months from completion of 1L therapy) and removed the need to have DLBCL relapse/progression after at least three months from completion of prior CD20 containing therapy; exclusion criterion 1b was updated to reflect this.

Source: Incyte, data on file (L-MIND CSR)(56)

Abbreviations: AE = adverse event; ADCC = antibody-dependent cellular cytotoxicity; ALT = alanine transaminase; AP = alkaline phosphatase; AST = aspartate aminotransferase; ASCT = autologous stem cell transplant; BCL = B-cell lymphoma; BSC = best supportive care; CNS = central nervous system; CR = complete response; DCR = disease control rate; DLBCL = diffuse large B-cell lymphoma; DoR = duration of response; EBV = Epstein Barr virus; ECOG = Eastern Cooperative Oncology Group; GCB = germinal centre B-cell; GEP = gastroenteropancreatic; HDC = high-dose chemotherapy; IMiD = immunomodulatory drug; IRC = independent radiology/clinical review committee; IV = intravenous; NK = natural killer; ORR = overall response rate; OS = overall survival; PET = positron emission tomography; PFS = progression-free survival; p.o. = taken orally; PR = partial response; R/R = relapsed or refractory; SD = stable disease; TAFA+LEN = tafasitamab + lenalidomide; TTNT = time-to-next treatment; TTP = time-to-progression; UK, United Kingdom; ULN = upper limit of normal;

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

US, United States; WHO = World Health Organization

The methodology of MOR208C201 study is summarised in Table 7.

Table 7. MOR208C201 methodology

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Source: Incyte, data on file (MOR208C201 CSR)(59)

Abbreviation: AE = adverse event; ALT = alanine transaminase; ANC = absolutely neutrophil count; AST = aspartate aminotransferase; ASCT = autologous stem cell transplant; BSC = best supportive care; CNS = central nervous system; CR = complete response; CT = computed tomography; DLBCL = diffuse large B-cell lymphoma; DoR = duration of response; ECOG = Eastern Cooperative Oncology Group; FDG-PET = [18F]-fluorodeoxyglucose-positron emission tomography; HIV = human immunodeficiency virus; IRC = independent radiology/clinical review committee; IV = intravenous; mAb = monoclonal antibody; MRI =

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

magnetic resonance imaging; NHL = non-Hodgkin’s lymphoma; NYHA = New York Heart Association; ORR = overall response rate; PFS = progression-free survival; PR = partial response; REAL = Revised European American Lymphoma; R/R = relapsed/refractory; SD = stable disease; TTP = time-to-progression; ULN = upper limit of normal; US = United States; WHO = World Health Organization.

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

B.2.4.1. Analysis population–L-MIND

In the L-MIND study, the following analysis populations were assessed(56):

• All patients screened: Consisted of all patients who signed informed consent and had a completed ‘informed consent’ electronic case report form (eCRF) page

• Enrolled patients: Consisted of all patients who received at least one dose of any study drug (tafasitamab or lenalidomide)

• Full analysis set (FAS): The FAS included all patients who received at least one dose of tafasitamab and at least one dose of lenalidomide. This meant that both study drugs had to be administered at least once. The FAS was the primary population for the analysis of efficacy and baseline characteristics. Of the 81 patients enrolled and treated in the study, one patient received tafasitamab only. Therefore, the FAS varied from 80 to 81 patients.

• Per protocol set (PPS): The PPS included all patients in the FAS who did not have any major protocol deviations that could confound the interpretation of the primary analyses conducted on the FAS. The PPS included all patients in the FAS who had received at least one dose of TAFA+LEN and underwent at least one post-baseline response assessment.

• Safety analysis set (SAS): The SAS included all patients who received at least one dose of tafasitamab or lenalidomide and had at least one post-baseline safety assessment. Valid safety assessments included documentation of death or a ‘no AE’ record. Analyses using the SAS were based on the study drug actually received.

• Pharmacokinetic analysis set (PKAS): The PKAS included all patients who received at least one dose of tafasitamab and had at least one quantifiable tafasitamab serum concentration.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

• Immunogenicity analysis set (IAS): The IAS included patients who had at least one anti-tafasitamab antibody assessment.

• Post-hoc: DLBCL FAS and DLBCL SAS: These consisted of the efficacy and safety analysis sets for the population with a centrally confirmed DLBCL diagnosis, used for a post-hoc analysis.

B.2.4.2. Analysis population–MOR208C201 study

In the MOR208C201 study, the following analysis populations were assessed (59):

• Intent-to-treat (ITT) population: Consisted of all patients who received at least one dose of study drug. Patients without any post-baseline assessment of NHL response were included as non-responders.

• Safety population: Consisted of all patients who received at least one dose of study drug

B.2.4.3. Statistical analyses

The primary and secondary endpoints in L-MIND were analysed descriptively for each analysis population using appropriate statistics (counts/percentages for discrete variables, mean, median, standard deviation, minimum, maximum, number of valid observations for continuous variables). For specific variables, p-values and 95% CIs were presented. No formal statistical hypothesis testing was planned.(56)

Similar to the L-MIND study, in the MOR208C201 study, endpoints were analysed descriptively.(59)

B.2.5. Quality assessment of the relevant clinical-effectiveness evidence

The quality assessments of L-MIND (Salles et al., 2020) and MOR208C201(59) are summarised in Table 8. Quality assessments of the studies identified by the SLR are summarised in Appendix D.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Table 8. Quality assessment for L-MIND (MOR208C203) and MOR208C201

Abbreviations: DLBCL = diffuse large B-cell lymphoma; FAS - full analysis set; IRC = independent radiology/clinical review committee; ITT = intention to treat; NA = not applicable; R/R = relapsed/refractory; SoC = standard of care

B.2.6. Clinical-effectiveness results of the relevant trials

B.2.6.1. L-MIND study (TAFA+LEN)

B.2.6.2. Patient disposition–L-MIND

In total, 156 patients were screened and 81 patients were enrolled in the L-MIND study. Overall, 30 (37.0%) completed the combination treatment phase on both study drugs (12 cycles). Patient disposition for L-MIND is summarised in Figure 5.

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

Figure 5. L-MIND study: patient disposition (all patients enrolled)

==> picture [411 x 145] intentionally omitted <==

==> picture [411 x 146] intentionally omitted <==

==> picture [411 x 145] intentionally omitted <==

Source: Incyte, data on file (L-MIND CSR)(56) Abbreviations: DLBCL = diffuse large B-cell lymphoma; HBV = hepatitis B virus; LEN = lenalidomide; MOR00208, tafasitamab; REAL = Revised European American Lymphoma; WHO = World Health Organization

As of the 30 October 2020 data cut-off, 19 patients remained on treatment and the median follow-up was 33.9 months (95% CI: 26.5, 35.4 months).(57)

Baseline characteristics–L-MIND

The L-MIND study enrolled a diverse group of patients, including difficult-to-treat subgroups, who represented patients treated in routine clinical practice. This suggested that the L-MIND study results would be reproducible in the real world.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Table 9 presents the baseline demographics and disease characteristics of the patients enrolled in L-MIND.

Table 9. L-MIND study: selected demographics and baseline characteristics

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

*Patients who were defined as primary refractory were excluded from the study. After a protocol revision, primary refractory disease was defined as disease progressing in the course of the 1L treatment as per International Working Group response criteria, and/or showing a response of less than a PR to 1L treatment or disease recurrence/progression within <6 months from the completion of 1L therapy. Note that an initial definition of

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

primary refractory DLBCL led to exclusion of relapses within three months of a prior anti-CD20 therapy. After revision, 15 patients in the L-MIND study (18.5%) were classified as having primary refractory disease.

†Defined as having a longest lesion diameter of ≥7.5 cm (by central assessment) ‡Patients without a PR or CR with salvage therapy or who had ASCT before enrolment

§**All patients who are not chemorefractory and who have comorbidities Other reasons include inability to successfully collect stem cells. Source: Salles et al., 2020.(53)

Abbreviations: ASCT = autologous stem cell transplantation; DLBCL = diffuse large B-cell lymphoma; ECOG = Eastern Cooperative Oncology Group; GCB, germinal centre B-cell; IHC, immunohistochemistry; IPI = International Prognostic Index; LDH = lactate dehydrogenase; PR, partial response; TAFA+LEN = tafasitamab + lenalidomide

Duration of treatment–L-MIND

In the primary analysis (30 November 2018 data cut-off), the median duration of exposure to study treatment (either TAFA+LEN or monotherapy with tafasitamab) in L-MIND was xxx months (interquartile range [IQR]: xxxxxxxx months). The median duration of exposure to lenalidomide was 6.2 months (IQR: xxxxxxxx months) and to tafasitamab monotherapy (following discontinuation of lenalidomide) was xxx months (IQR: xxxxxxxx months).(53)

After the 30 October 2020 data cut-off, the median duration of exposure to study treatment (either TAFA+LEN or monotherapy with tafasitamab), was 9.2 months (range: xxx months–xxxx months).(54) The median duration of exposure to lenalidomide was xxx weeks (range: xxx weeks–xxxx weeks).(57) The median duration of exposure to monotherapy with tafasitamab after lenalidomide discontinuation was xxxx months (range: xxxxxxxx months).(57)

B.2.6.3. MOR208C201 study (tafasitamab monotherapy)

Patient disposition–MOR208C201 DLBCL cohort

Fourteen patients with DLBCL were enrolled in stage 1 of the MOR208C201 study; this DLBCL cohort was expanded to 35 patients in stage 2. Ten patients in the DLBCL cohort discontinued the study (five from progressive disease, three died, one was withdrawn by the investigator, and one discontinued due to a protocol violation).(60)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Twenty-five patients completed the study and 12 continued to cycle 3. Six patients remained on maintenance treatment and one patient remained at the date of data cut-off, 28 September 2018.(60)

Baseline characteristics– MOR208C201 DLBCL cohort

Table 10 presents the baseline demographics and disease characteristics of the patients enrolled in MOR208C201.

Table 10. MOR208C201 study: selected demographics and baseline characteristics– DLBCL cohort (FAS)

Source: Incyte, data on file (MOR208C201 CSR)(61) Abbreviations: DLBCL = diffuse large B-cell lymphoma; ECOG = Eastern Cooperative Oncology Group

B.2.6.4. Efficacy outcomes in L-MIND

At the time of writing, the latest data cut available for the L-MIND study is the third planned interim analysis, with follow-up of ≥35 months (data cut-off 30 October

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

2020). The clinical benefit of TAFA+LEN in patients with R/R DLBCL followed the same trend observed in the early interim analyses.

Primary efficacy outcomes–L-MIND

Nineteen of 22 patients who were receiving ongoing tafasitamab treatment were assessed through new tumour imaging and/or clinical data accumulated between the data cut-offs of 30 November 2019 and 30 October 2020. For 15 patients, the best response did not change. For two patients, the best response changed from PR to CR, and for two additional patients, the best response changed from CR to PR. The best objective response was CR for 32 patients (n=32/80; 40%) and PR for 14 patients (n=14/80; 18%). Based on these data, the best ORR as assessed by independent radiology/clinical review committee (IRC) was 57.5% (95% CI: 45.9%, 68.5%),(57) consistent with analyses at the previous data cut-offs.

Twenty-six patients had stable disease or progressive disease (PD; n=13/80; 16.3% for each group) as their best objective response. As in the initial analysis, eight (n=8/80; 10.0%) patients were not evaluable, as no valid post-baseline radiological examination for response assessment was available, or the baseline scan was inadequate. These patients were included as non-responders in the analysis.(57) The best ORR data at this timepoint are summarised in Table 11.

Table 11. Best ORR (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

*CR + PR

Source: Source: Incyte, data on file (L-MIND CSR Addendum 3)(57) Abbreviations: CI = confidence interval; CR = complete response; ORR = objective response rate; PD = progressive disease; PR = partial response; SD = stable disease; TAFA+LEN = tafasitamab + lenalidomide

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Secondary efficacy outcomes–L-MIND

Duration of response

As of the 30 October 2020 data cut-off, the median duration of response (DoR) was 43.9 months (95% CI: 26.1, not reached). Of the 46 responders, 13 (n=13/80; 28.3%) patients progressed, two (n=2/80; 4.3%) patients died, and 31 (n=31/80; 67.4%) patients were censored. Figure 6 shows the Kaplan-Meier (KM) plot for patients in the FAS. A KM probability estimate for DoR at 12 months was 73.7% (95% CI: 57.4%, 84.5%), at 18 months was xxxxxxxxxxxxxxxxxxxxxx, at 24 months was xxxxxxxxxxxxxxxxxxxxxx, and at 30, 36, and 42 months was

xxxxxxxxxxxxxxxxxxxxxx.(57) These long-term data further demonstrated that a

durable response was achieved in a substantial proportion of patients receiving TAFA+LEN.

Figure 6. KM plot of DoR (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

==> picture [447 x 223] intentionally omitted <==

Note: In case the median or the respective confidence limits were not calculable by the KM method, NR is displayed instead. Source: (57)

Abbreviations: CI = confidence interval; LEN = lenalidomide; NR = not reached

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

A KM plot of DoR by best objective response CR or PR for patients in the FAS (IRC evaluation) is presented in

Figure 7.(57) Of the xx patients with a best objective response of CR, xxxxx patients progressed (xxxxxxxxxx), xxx patient died (xxxxxxxxxx), and xx patients were censored (xxxxxxxxxxxxx).(57) The estimate of the median DoR for patients with a best objective response of CR was not reached. The KM probability estimate for patients with a best objective response of CR was

xxxxxxxxxxxxxxxxxxxxxxx at 12 months, xxxxxxxxxxxxxxxxxxxxxxxxx at 18 months, xxxxxxxxxxxxxxxxxxxxxxxxxxx at 24 months, and xxxxxxxxxxxxxxxx at

30, 36 and 42 months.(57) Of the xx patients (xxxxxxxxxx) with PR, the median

DoR was xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx; xxxxx patients progressed (xxxxxxxxxxxxx), xxx patient died (xxxxxxxxxxxxx), and xxxx patients were censored (xxxxxxxxxxx).(57)

Figure 7. KM plot of DoR by best objective response (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

==> picture [440 x 118] intentionally omitted <==

==> picture [440 x 117] intentionally omitted <==

Notes: In case the median or the respective confidence limits were not calculable by the KM method, NR is displayed instead. The 34 patients with best objective response not PR or CR were not included in this subgroup analysis.

Source: (57)

NOTE: Permission must be sought from the publisher before reproducing this figure for use with an external audience.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Abbreviations: CI = confidence interval; CR = complete response; NR = not reached; PR = partial response

PFS

PFS data provided additional support for the efficacy and durable responses demonstrated by the ORR and DoR data, with consistent results achieved at each data cut-off.

PFS events were observed in 42 patients (n=42/80; 52.5%). A KM curve of PFS in the FAS is presented in Figure 8. The KM estimate for the median PFS was 11.6 months (95% CI: 6.3, 45.7 months) with a median follow-up time of 33.9 months (xxxxxxxxxxxxxxxxxxx).(57)

Figure 8. KM plot of PFS (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

==> picture [452 x 269] intentionally omitted <==

Note: In case the median or the respective confidence limits were not calculable by the KM method, NR is displayed instead.

Source: (57)

Abbreviations: CI = confidence interval; LEN = lenalidomide; NR = not reached

Patients continued to receive a PFS benefit from tafasitamab monotherapy after the combination treatment period had ended and lenalidomide had been discontinued. A

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

post-hoc analysis at the 30 November 2018 data cut-off (median follow-up 17.3 months [95% CI: 11.5, 21.2 months]) showed that the median PFS was 12.7 months (95% CI: 2.3 months, not reached) after discontinuation of lenalidomide (while still on tafasitamab monotherapy).(38) Furthermore, PFS was longer in patients receiving 2L vs ≥3L treatment: 23.5 months (95% CI: 7.4 months, not reached) and 7.6 months (95% CI: 2.7 months, not reached) respectively.(54)

Time to progression and time-to-next treatment

The median time to progression (TTP) was 16.2 months (95% CI: 17.4 months, not reached) and PFS events occurred in 35 of 80 patients (44%). The median time-tonext treatment (TTNT) was 15.4 months (95% CI: 7.6 months, not reached) and 43 of 80 patients (54%) received subsequent treatment. Two patients subsequently received salvage treatment consolidation with stem cell transplant (one patient each with ASCT and allogeneic stem cell transplant). One other patient subsequently received CD19 CAR-T therapy after disease progression, had a CR, and was in remission at the time of data cut-off (30 November 2018).(53)

OS

The KM estimate for median OS was 33.5 months (95% CI: 18.3 months, not reached; FAS; Figure 9) with a median follow-up time of 42.7 months (95% CI: 38.0, 47.2 months).(57) Overall, 41 patients died (n=41/80; 51.3%). Thirty-nine patients were censored in the OS analysis, including one patient censored due to being lost to OS follow-up. The KM probability estimate of OS at 12 months was

xxxxxxxxxxxxxxxxxxxxxxxxxxx), xxxxxxxxxxxxxxxxxxxxxxxxxxx) at 18 months, xxxxxxxxxxxxxxxxxxxxxxxxxxxx at 24 months, xxxxxxxxxxxxxxxxxxxxxxxxxxxx at 30 months,

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxx.(57)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Figure 9. KM plot of OS (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

==> picture [452 x 268] intentionally omitted <==

Note: In case the median or the respective confidence limits were not calculable by the KM method, NR is displayed instead.

Source:(57)

Abbreviations: CI = confidence interval; LEN = lenalidomide; NR = not reached

The KM estimate for median OS by best objective response of CR (IRC) was not reached (95% CI: 45.7 months, not reached; FAS; Figure 10) at the 30 October 2020 cut-off date.(57) For this subgroup, the KM probability estimate of OS was 96.9% (95% CI: 79.8%, 99.6%) at 18 months, 90.6% (95% CI: 73.7%, 96.9%) at 24 months, 81.3% (95% CI: 62.9%, 91.1%) at 36 months, and

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.(57) The KM estimate for median OS by best objective response of PR was 22.5 months (95% CI: 8.5 months, not reached; FAS; Figure 10).

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

Figure 10. KM plot of OS by best objective response (updated analysis data cut-off 30 October 2020; FAS; IRC assessed)

==> picture [430 x 235] intentionally omitted <==

Note: In case the median or the respective confidence limits were not calculable by the KM method, NR was displayed instead. Thirty-four patients with best objective response not PR or CR were not included in this subgroup analysis.

Source: (57)

Abbreviations: CI = confidence interval; CR = complete response; NR = not reached; PR = partial response

Additional outcomes–L-MIND

Time to response and time to CR; primary analysis

The depth of the response achieved with TAFA+LEN was supported by the time to response and time to CR data. Time to response was defined as the date of assessment of first documented response of CR or PR minus the date of first administration of any study drug.(57) In the primary analysis, the median time to response was 2.1 months (IQR: xxxxxxxxx).(54, 57) Time to CR was defined as the date of assessment of first documented response of CR minus the date of first administration of any study drug.(57) In the primary analysis, the median time to CR was 6.8 months (IQR: xxxxxxxxxx).(54, 57)

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

Patients with a c-MYC translocation; primary analysis

A post-hoc evaluation of the 30 November 2018 primary analysis showed that seven patients had a c-MYC translocation that was identified during central pathology review: of these patients, three had a CR, one had a PR, and three did not respond to therapy. Another of the seven patients presented with a double-hit translocation and had a PR (lasting 5.8 months); one more had a triple-hit translocation and a CR (ongoing at data cut-off: 20.1 months).(53) c-MYC translocations, particularly in combination with one or more additional mutations (i.e., double- or triple-hit disease), are associated with a high risk of progression and poor outcomes.(27)

Patients with central pathology-confirmed DLBCL

Efficacy analyses based on the data cut-off of xxxxxxxxxxxxxxx were evaluated in patients who had both a local pathology and a central pathology histological diagnosis of DLBCL. xxxxxxxxxxxxx patients were included in the DLBCL primary efficacy SAS andxxxxxxxxxxxxx patients were included in the DLBCL SAS).(57)

The distribution of patients in subgroups of prognostically important covariates were comparable between the full population and the patients with centrally confirmed DLBCL. Therefore, no imbalance in baseline factors was present that could confound the interpretation of efficacy results between the full population (FAS, n=xxxor SAS, n=xx) and the subgroup of patients with centrally confirmed DLBCL (based on DLBCL FAS, n=xx or DLBCL SAS, n=xx).(57)

The following efficacy outcomes were reported for the DLBCL FAS population as of the xxxxxxxxxxxxxxxxdata cut-off date:

• Best objective response was CR for xxxxxxxxxxxxx patients and PR for xxxxxxxxxxxxxx patients. Based on these data, the IRC-assessed best ORR was xxxxxxxxxxxxxxxxxxxxxxxxxx.(57)

• The estimate for the median DoR was

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx. Of the xxxresponders,

xxxxxxxxxx patients progressed, xxxxxxxxxxxpatients died, and xxxxxxxxxx patients were censored.(57)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

• The median follow-up time for PFS was xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx in the DLBCL FAS. The KM estimate for median PFS was

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx.(57)

• The median follow-up time for OS was xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx The KM estimate for the median OS was

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Efficacy results between the investigator (INV)-confirmed DLBCL and patients with central pathology-confirmed DLBCL were consistent, and estimates for primary and secondary endpoints (i.e., best ORR, CR rate, median DoR, median PFS and median OS) between patients with DLBCL as per INV and as per central pathology were comparable, supporting the interpretation of efficacy results based on the primary efficacy population (n=xx patients).(57) This reflected routine clinical practice where patients are diagnosed locally without central reassessment. The validity of this approach also confirmed when the EMA approved tafasitamab based on a review of the L-MIND results. xxxxxxxxxxxxxxxxxxx patients who were still on treatment at the latest data cut-off had their initial DLBCL diagnosis confirmed centrally.

B.2.6.5. Efficacy outcomes in MOR208C201

Primary efficacy outcomes–MOR208C201 DLBCL cohort

The primary efficacy analysis for the DLBCL cohort is summarised in Table 12. The data cut-off for the primary analysis was 28 September 2018. The median duration of exposure to tafasitamab in the DLBCL cohort was 7.1 weeks (range: 0–232 weeks). The ORR was 25.7% (95% CI: 12.5%, 43.3%), demonstrating the efficacy of tafasitamab single-agent treatment in this population.(61)

Table 12. Primary efficacy analysis: MOR208C201 (ITT population)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Abbreviations: CI = confidence interval; CR = complete response; DCR = disease control rate; ORR = overall response rate; NE = not estimable; PD = progressive disease; PR = partial response; SD = stable disease Source: Incyte, data on file (MOR208C201 CSR)(61)

B.2.7. Subgroup analysis

Subgroup analyses are presented in Appendix E.

B.2.8. Meta-analysis

A meta-analysis study is not presented as part of the clinical evidence.

B.2.9. Indirect and mixed treatment comparisons

As the pivotal L-MIND study of TAFA+LEN in R/R DLBCL (Section B.2. ) was a single-arm trial, the comparative efficacy of TAFA+LEN was assessed via 1:1 nearest-neighbour (NN) matching with external (synthetic) control arms. These data were generated from two generated in two retrospective cohort studies (RE-MIND [MOR208C206] and RE-MIND2 [MOR208C213]),(62, 63) and a matching-adjusted indirect comparison (MAIC) against the published clinical studies of key comparators.(64)

In line with the final decision problem, the comparators considered most relevant to the UK market according to expert clinical opinion are listed below.

 Pola-BR

• Gemcitabine, oxaliplatin (R-GemOx)

• BR

Data for these comparisons were provided by RE-MIND2 and the MAIC (Sections B.2.9.2. and Appendix D), while the RE-MIND comparison of TAFA+LEN vs.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

lenalidomide monotherapy provided additional evidence regarding the efficacy and potential synergy of TAFA+LEN compared with lenalidomide monotherapy, and is described briefly for context in Section B.2.9.1.

B.2.9.1. RE-MIND

The RE-MIND study was an estimated propensity-score (ePS)-based 1:1 NN matched comparison, designed to quantify the additional benefit of combining tafasitamab with lenalidomide.(62) Details of the methodology and patient population can be found in Appendix D.

A statistically significant improvement was seen with TAFA+LEN vs. lenalidomide monotherapy in endpoints including: best ORR (67.1% [95% CI, 55.4%, 77.5%] vs. 34.2% [95% CI, 23.7%, 46.0%]; odds ratio [OR]: 3.9 [95% CI: 1.9, 8.1]; p<0.0001); median PFS (12.1 months vs. 4.0 months; hazard ratio [HR]: 0.463 [95% CI: 0.307, 0.698]; p=0.0002); and median OS (not reached vs. 9.3 months; HR: 0.499 [95% CI: 0.317, 0.785]; p=0.0026).(62) Key efficacy endpoints are summarised in Table 13. The best ORR was defined as the proportion of patients with CR or PR as best response achieved at any time within the analysis window (index to 32 months (974 days) or between index date and date of initiation of a new anti-DLBCL medication or death. The denominator was the total number of patients included in the analysis set.

Table 13. RE-MIND study: overview of efficacy outcomes–MAS25

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Notes:

1Clopper-Pearson exact method

2Logistic regression for unpaired data; logistic regression model: response=cohort status

3Cox proportional hazard model

Source: Incyte, data on file (RE-MIND CSR)(62)

Abbreviations: CI = confidence interval; CR = complete response; DCR = disease control rate; EFS = event-free survival; HR = hazard ratio; LEN = lenalidomide; NR = not reached; OR = odds ratio; ORR = objective response rate; PD = progressive disease; PFS = progression-free survival; PR = partial response; SD = stable disease; SE = standard error; TAFA+LEN = tafasitamab + lenalidomide; TTNT = time-to-next treatment

Although L-MIND demonstrated the clear benefit of adding tafasitamab to

lenalidomide monotherapy for the management of R/R DLBCL, clinical experts from the UK highlighted{Incyte Corporation, 2020 #316} that, as lenalidomide

monotherapy is not frequently used to treat R/R DLBCL in clinical practice, the REMIND study was not relevant for assessing comparative efficacy in the UK clinical practice setting. Additional data from RE-MIND, including sensitivity analyses, can be found in Appendix D.1.4.9.

B.2.9.2. RE-MIND2

RE-MIND2 was a large, real-world, retrospective cohort study of patients with R/R DLBCL (N=3,454) ), based on a pre-specified design, aimed at characterising the effectiveness and tolerability of TAFA+LEN (in L-MIND; data cut-off 30 October 2020) with a 1:1 NN-matched population treated with systemic regimens administered in routine clinical care as recommended by NCCN/ESMO guidelines.(63) The RE-MIND2 cohort included patients treated with the following regimens: BR, R-GemOx, pola-BR, rituximab (R)+lenalidomide (LEN), CAR-T therapies, and pixantrone; in the second, third, or fourth-line treatment settings.(63)

Based on feedback from UK clinical experts{Incyte Corporation, 2020 #316}, BR, R-GemOx and pola-BR were considered the most relevant comparators for patients with R/R DLBCL who are ineligible for ASCT in the UK; therefore, the RE-MIND2 study provided a relevant and meaningful comparison of outcomes for TAFA+LEN

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

against therapies used in current UK clinical practice. Overall, there were five UKbased patients enrolled into L-MIND (all of whom received TAFA+LEN) and 115 UKbased patients enrolled into Re-MIND2 (receiving different systemic therapies).(63)

RE-MIND2 methodology overview

A cohort of 3,454 patients was selected from sites in Europe, North America and the Asia-Pacific region according to the inclusion and exclusion criteria outlined in Table 14. Several key eligibility criteria were identical to those employed in L-MIND to enable comparison between populations.(63).

Table 14. Inclusion and exclusion criteria for the RE-MIND2 study

• Inclusion criteria Exclusion criteria  ≥18 years at initial DLBCL diagnosis  Patients with CNS involvement by lymphoma at initial DLBCL diagnosis

• One of the following histologically-confirmed diagnosis:

  • Patients who were treated with CD19-targeted therapy or immunomodulatory drugs as a frontline DLBCL therapy

• DLBCL not otherwise specified

• T-cell/histiocyte rich large B-cell lymphoma

  • Patients who underwent an allogeneic stem cell transplant

• EBV-positive DLBCL of the elderly

• Grade 3b follicular lymphoma

  • Patients who had prior history of malignancies other than DLBCL, unless the patient has been free of the disease for ≥5 years prior to inclusion. Exceptions to this time limit include a history of the following:

• Composite lymphoma with a DLBCL component with a subsequent DLBCL relapse (according to REAL/WHO) classification

• Evidence of histological transformation to DLBCL from an earlier diagnosis of low-grade lymphoma with a subsequent DLBCL relapse

  • Basal cell carcinoma of the skin

  • Squamous cell carcinoma of the skin

• R/R DLBCL and received at least two systemic regimens for the treatment of DLBCL, including at least one anti-CD20 containing therapy

  • Carcinoma in situ of the cervix

  • Carcinoma in situ of the breast

  • Carcinoma in situ of the bladder

  • Incidental histological finding of prostate cancer (Tumour/Node/Metastasis stage of T1a or T1b)

• Patients who received tafasitamab.

• Patients who were human immunodeficiency virus positive (applicable to sites in Taiwan only).

Source: Incyte, data on file (RE-MIND2 CSR)(63)

Abbreviations: CNS = central nervous system; DLBCL = diffuse large B-cell lymphoma; EBV = Epstein-Barr virus; REAL = Revised European American Lymphoma; R/R = relapsed or refractory; WHO = World Health Organization

In addition, the non-randomised cohorts were balanced with the L-MIND population on nine baseline covariates using estimated propensity score (ePS; Table 15), with additional sensitivity analyses conducted with matching according to 11 covariates.(63)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Table 15. Baseline covariates used in the ePS for RE-MIND2

Baseline covariates Age (as categorical variable with subgroups <70 vs. ≥ 70 years of age) Ann Arbor stage (I/II vs. III/IV) Refractoriness status to last therapy line (yes vs. no) Number of prior lines of therapy (1 vs. 2/3) History of primary refractoriness (yes vs. no) Prior ASCT (yes vs. no) Neutropenia (<1.5×109/L) (yes vs. no) Anaemia (<10 g/dL [=6.21 mmol/L]*) (yes vs. no) Elevated LDH (LDH>upper limit of normal [ULN]) (yes vs. no)

*Conversion formula (g/dL×0.621=mmol/L) Source: Incyte, data on file (RE-MIND2 CSR)(63)

Abbreviations: ASCT = autologous stem cell transplant; LDH = lactate dehydrogenase; ULN = upper limit of normal

Data from the L-MIND study database (data cut-off 30 November 2019; i.e.,

approximately two years after the last patient was enrolled in the study)(53) were compared with the following observational cohorts in RE-MIND2(63) (key comparators are highlighted in bold):

• Systemic therapies pooled cohort

• BR cohort

• R-GemOx cohort

• R + LEN (R2) cohort

• CD19 CAR-T cohort (pre-specified sensitivity analysis)

• Pola-BR cohort (pre-specified sensitivity analysis)

• Pixantrone monotherapy cohort

The high degree of cohort-balancing using ePS-based 1:1 matching allowed for a more robust estimation of treatment effect between the TAFA+LEN cohort and the primary analysis cohorts of systemic therapies pooled, BR, and R-GemOx than would have been afforded by other balancing methods.(63).

In the L-MIND study, the administration of TAFA+LEN was followed by tafasitamab monotherapy until disease progression,(53) whereas other comparator therapies in RE-MIND2 were administered for a fixed duration. The analysis window for

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

observational cohorts was therefore defined as the interval between the index date for the given treatment line plus 44 months (1,338 days).(63) Key study endpoints are listed below; a full list of endpoints and subgroup analyses is provided in Appendix D.5.3. (63)

• Primary endpoint: OS

• Secondary endpoints:

  •  ORR

  •  CR rate

  •  DoR

  • Event-free survival (EFS)

  •  PFS

  •  TTNT

  •  Treatment discontinuation rate due to AEs

  • Duration of treatment exposure

RE-MIND2 primary analysis results

OS

The difference in OS between cohorts was statistically significant in favour of TAFA+LEN vs. BR (HR=0.418 [95% CI: 0.272, 0.644]; Cox proportional hazard model p<0.0001; Figure 11), and R-GemOx (HR=0.467 [95% CI: 0.305, 0.714]; Cox proportional hazard model p=0.0004; Figure 11).(63) Thus, the RE-MIND2 study met its primary endpoint, showing statistically significant improvements in OS for TAFA+LEN vs. BR, R-GemOx and the pooled cohort of all systemic therapies for R/R DLBCL listed in the NCCN/ESMO guidelines.(63) Clinical expert opinion aligned with the OS for the comparators of R-GemOx and BR, although BR was not a commonly used regimen.{Incyte Corporation, 2020 #316}

The proportion of patients who had an OS event in the TAFA+LEN cohort was lower with TAFA+LEN compared with BR (48.0% vs. 70.7%), and R-GemOx (48.6% vs. 74.3%). The main cause of the OS event was DLBCL disease progression in all the cohorts. The median OS (KM estimate) was longer in the TAFA+LEN cohort

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

compared with BR (31.6 vs. 9.9 months), and R-GemOx (31.6 vs. 11.0 months). The probability of patients surviving at month 12 was 74.0% in the TAFA+LEN cohorts and was 41.4% and 44.7% in the BR and R-GemOx cohorts, respectively.(63)

Figure 11. KM plot for OS: BR (a) and R-GemOx (b)

(a) BR

==> picture [451 x 121] intentionally omitted <==

==> picture [451 x 121] intentionally omitted <==

==> picture [451 x 121] intentionally omitted <==

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

(b) R-GemOx

==> picture [472 x 128] intentionally omitted <==

==> picture [472 x 127] intentionally omitted <==

==> picture [472 x 128] intentionally omitted <==

Notes: MAS_Pool included 1:1 matched patients from the L-MIND study and the observational cohort using nine baseline covariates. MAS_BR included 1:1 matched patients from the L-MIND study and BR as pre-specified treatment. MAS_R-GemOx included 1:1 matched patients from the L-MIND study and R GemOx as pre-specified treatment.

The median was calculated with the KM method. The 95% CI was calculated by means of Greenwood formula. HR was calculated with Cox proportional hazard model.

Source: Incyte, data on file (RE-MIND2 CSR)(63)

Abbreviations: BR = bendamustine and rituximab; CI = confidence interval; KM = Kaplan-Meier; MAS, matched analysis set; NR = not reached; R-GemOx = rituximab in combination with gemcitabine, oxaliplatin; TAFA+LEN = tafasitamab + lenalidomide

A forest plot of OS HRs with 95% CIs using the Cox proportional hazard model for

the different analysis sets is provided in Figure 12.

Figure 12. Forest plot of OS HRs with 95% CIs using Cox proportional hazard model for different analysis sets

(a) Matched analysis set for TAFA+LEN vs. BR

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

==> picture [453 x 121] intentionally omitted <==

(b) Matched analysis set for TAFA+LEN vs. R-GemOx

==> picture [453 x 120] intentionally omitted <==

Notes: HR was calculated using the observational cohort as reference cohort. HR <1.0 favours TAFA+LEN. Source: Incyte, data on file (RE-MIND2 CSR)(63)

Abbreviations: BR = bendamustine and rituximab; CI = confidence interval; E/E = number of events in TAFA+LEN/the observational cohort; MAS = matched analysis set; R-GemOx = rituximab in combination with gemcitabine, oxaliplatin; TAFA+LEN = tafasitamab + lenalidomide

The subgroup analysis by age of OS was consistent with the primary matched analysis results (Appendix D). Outcomes were broadly consistent across the different matching analyses for each comparison, with improvements in OS for TAFA+LEN in each case.

Summary of secondary efficacy endpoints

All time-to-event endpoints (PFS, EFS, and TTNT) supported the primary analysis results of OS and aligned with the overall results, with differences between the TAFA+LEN cohort and the BR, and R-GemOx cohorts. The median PFS in the TAFA+LEN cohort was longer compared with the cohorts of systemic therapies pooled (12.1 vs. 4.6 months) and R-GemOx (9.1 vs. 4.0 months). The median (KM estimate) PFS in the BR cohort was longer compared with the TAFA+LEN cohort (11.5 vs. 8.7 months). Moreover, a significantly higher ORR was observed in the

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

TAFA+LEN cohort compared with the cohorts of systemic therapies pooled and R- GemOx.

A detailed overview of the secondary efficacy endpoints is presented in Appendix D.

Summary of safety endpoints

Treatment discontinuation rate due to AEs

AEs leading to permanent discontinuation of treatment were reported for 156 (4.7%) patients in the Ob-ENR[1] analysis set. Eight patients discontinued due to AEs in the TAFA+LEN cohort compared with BR (14.5%) and R-GemOx (15.1%). In the BR, and R-GemOx cohorts, two (2.8%), and four (5.4%) patients, respectively, had AEs leading to permanent discontinuation of treatment.(63)

The sensitivity analysis confirmed the primary analysis with seven (14.9%), and six (13.6%) patients who discontinued due to the AEs in the TAFA+LEN cohort for BR, and R-GemOx, respectively. The number of patients who discontinued due to the AEs in the BR and R-GemOx cohorts were three (4.8%), and one (1.7%), respectively.(63) The longer exposure in the TAFA+LEN cohort also indicated a favourable tolerability profile of this regimen.(63)

Duration of treatment exposure

The median duration of exposure in the TAFA+LEN cohort was longer (xxxxxxxxxxxxxx months) compared with BR (xxx months) and R-GemOx (xxx months). This difference can be attributed to the respective treatment regimens. In the L-MIND study, the administration schedule for TAFA+LEN was 12 cycles (approximately 12 months), followed by tafasitamab monotherapy until disease progression. In comparison, most therapies administered in the BR and R-GemOx cohorts were immunochemotherapies, which are typically administered over a fixed, limited treatment duration of approximately two to six months.

1 The Ob-ENR included all patients enrolled in the observational study.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

RE-MIND2 – additional analyses

Sensitivity analyses confirmed the main analyses of the primary (i.e., OS) and secondary endpoints. Details of the sensitivity analyses are provided in Appendix D.

RE-MIND2 included pre-specified exploratory of TAFA+LEN vs. pola-BR and vs. CAR-T therapy. Due to the recent approvals of both therapies, there were insufficient patient numbers with which to conduct the 1:1 NN matching analysis.(63) Instead 1:1 matching was undertaken based on 9 covariates with multiple imputations.

The TAFA+LEN cohort exhibited xxxxxxxxx survival time than the pola-BR cohort, with a median survival time of xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx compared with xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx in the pola-BR

cohort. The improvement was statistically significant using propensity score analysis (HR, 0.42 [0.23, 0.78]; xxxxxxxx). xxxxxxxxxxxx individuals in each cohort experienced PD xxxxxxxin the TAFA+LEN cohort, xxxxxxin the pola-BR cohort); approximately xxxxof subjects in both cohorts died xxxxxxxin TAFA+LEN vs. - xxxxxxxn pola BR). The median progression event time (KM estimate) was

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxfor the TAFA+LEN cohort and xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx for the pola-BR cohort. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxx p-value =

xxxxxx; HR =xxxxxx with Cox proportional hazard model p-value = xxxxxxxxx

B.2.9.3. MAIC

In the absence of head-to-head clinical studies of TAFA+LEN vs. comparators, an indirect treatment comparison was designed to evaluate the relative efficacy of TAFA+LEN in L-MIND vs. published comparator studies, including pola-BR, BR and R-GemOx. The population from L-MIND was matched with the published comparator populations via an MAIC.

MAIC methodology overview

Six prospective studies were selected for inclusion in the MAIC (

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Table 16). The studies were selected based on an SLR and interviews with clinical experts, to enable a meaningful, population-adjusted comparison to the L-MIND study.{Incyte Corporation, 2020 #316} The MAICs were conducted using the methods described by Signorovitch et al., 2012(65) following current NICE guidelines.(66) For further details on the identification of studies and methodology for the MAIC, and a full list of studies identified in the SLR (and reasons for exclusion where relevant), please see Appendix D.

Three studies reporting data for BR were included in the MAIC: the GO29365 trial of pola-BR vs. BR,(21, 40) the Vacirca et al., 2014 study,(67) and the Ohmachi et al., 2013 study.(68) An MAIC comparing L-MIND with pooled BR cohort data from the three trials was also conducted with further details included in Appendix D.

There were conflicting estimates of response rates observed for the GO29365 trial. The Sehn et al., Journal of Clinical Oncology paper(21) reported 25 patients with CR or PR according to the IRC, while the Sehn et al., paper(21) only mentions 19 patients with IRC-CR or IRC-PR. As the breakdown of patients by response type was not reported in this source, these data were not investigated further.

The United States (US) Food and Drug Administration (FDA) re-analysis of the GO29365 trial explicitly censored PFS records of patients who received a subsequent anti-cancer treatment without a recorded progression events at the time of the last progression assessment available. A similar censoring rules was used in the L-MIND study, and as a result, the PFS reported by the FDA re-analysis appeared more comparable to the L-MIND data than the PFS reported in the Sehn et al., Journal of Clinical Oncology paper. Therefore, the comparative analyses against the data reported in the FDA dossier were used in the base-case analyses. Comparative analyses for PFS-IRC used the Sehn et al., Journal of Clinical Oncology paper as a data source.

Table 16. Studies identified for the MAIC study by the SLR and clinician interviews

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

aThere were conflicting estimates of response rates observed for the GO29365 trial. The Sehn et al., Journal of Clinical Oncology paper reported 25 patients with CR or PR according to the IRC, while the Sehn et al., Blood paper only mentions 19 patients with IRC-CR or IRC-PR. As the breakdown of patients by response type was not reported in this source, these data were not investigated further.

bThe FDA re-analysis of the GO29365 trial explicitly censored PFS records of patients who received a subsequent anti-cancer treatment without a recorded progression events at the time of the last progression assessment available. A similar censoring rules was used in the L-MIND study, and as a result, the PFS reported by the FDA re-analysis appeared more comparable to the L-MIND data than the PFS reported in the Sehn et al., Journal of Clinical Oncology paper. Therefore, the comparative analyses against the data reported in the FDA dossier were used in the base-case analyses. Comparative analyses for PFS-IRC used the Sehn et al., Journal of Clinical Oncology paper as a data source.

Abbreviations: BR = bendamustine and rituximab; CI = confidence interval; CRR = complete response rate; DoR = duration of response; FDA = Food and Drug Administration; IRC = independent radiology/clinical review committee; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; pola-BR = polatuzumab, bendamustine, and rituximab; R-GemOX = rituximab, gemcitabine, oxaliplatin

MAIC results

Results of the MAIC vs. the key comparator cohorts of pola-BR and BR are presented below; results vs. lenalidomide monotherapy and R-GemOx are presented in Appendix D. An overview of the relative efficacy estimates for TAFA+LEN compared with all comparators (pola-BR, BR) across all efficacy outcomes is also provided in Appendix D. The best response changed for some patients during reassessment between the xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx data cuts, which accounts for some small differences in patient numbers.

Matching scenarios and baseline characteristics

Details of the matching scenarios and baseline characteristics for the key comparators pola-BR and BR, compared with the L-MIND observed and matched populations for each MAIC analysis, are included in Appendix D. Successful matching was achieved for all three comparators, allowing meaningful assessment of the relative efficacy of TAFA+LEN in L-MIND vs. each of the three comparators.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

OS

The estimated HRs for TAFA+LEN for the unadjusted and adjusted L-MIND populations vs. comparators are shown in Appendix D. KM curves for the L-MIND adjusted and unadjusted populations vs. comparators are shown in Figure 13.

Figure 13. KM estimates for OS for TAFA+LEN observed (green) and adjusted (blue) compared with reported OS estimates for comparators (red)

MAIC vs. pola-BR

==> picture [266 x 190] intentionally omitted <==

MAIC vs. BR (GO29465 study)

==> picture [265 x 185] intentionally omitted <==

Source: MAIC technical report(64)

Abbreviations: BR = bendamustine and rituximab; MAIC = matching-adjusted indirect comparison; pola-BR = polatuzumab vedotin with bendamustine and rituximab

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Results for the assessment of the proportionality of hazard assumption for each MAIC are presented in Appendix D. No concerns were raised regarding the assessments for OS vs. R-GemOx. In the MAIC vs. BR (GO29365 study), the distance between the TAFA+LEN and BR curves increased over time, hinting at a potential violation of the proportional hazards assumption.

PFS

The estimated HRs for PFS with TAFA+LEN in the unadjusted and adjusted L-MIND populations vs. comparators are shown in Appendix D. KM curves for the L-MIND adjusted and unadjusted populations vs. comparators are shown in Figure 14.

Figure 14. KM estimates for PFS for TAFA+LEN observed (green) and weighted (blue) compared with reported PFS-IRC estimates for comparators (red)

MAIC vs. pola-BR (PFS-IRC)

==> picture [220 x 77] intentionally omitted <==

==> picture [220 x 77] intentionally omitted <==

MAIC vs. BR (PFS-IRC)

==> picture [452 x 166] intentionally omitted <==

----- Start of picture text -----
GO29365 Vacirca et al., 2014
----- End of picture text -----

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Ohmachi et al., 2013

==> picture [223 x 153] intentionally omitted <==

Pooled naïve and MAIC-adjusted estimates of HRs

==> picture [181 x 225] intentionally omitted <==

Source: MAIC technical report(64)

BR = bendamustine and rituximab; INV = investigator-assessed; IRC = independent review committee; KM = Kaplan-Meier; PFS = progression-free survival; R-GemOx = rituximab in combination with gemcitabine, oxaliplatin

In the MAIC vs. BR (GO29365 and Vacirca et al., studies), no major concerns were identified. Although the TAFA+LEN and BR curves were observed to overlap initially for PFS-IRC in GO29365 and Vacirca et al., they quickly separated.

DoR

The estimated HRs for DoR with TAFA+LEN in the unadjusted and adjusted L-MIND populations vs. comparators are shown in Appendix D. KM curves for the L-MIND adjusted and unadjusted populations vs. comparators are shown in

Figure 15.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Figure 15. KM estimates for DoR for TAFA+LEN observed (green) and weighted (blue) compared with reported DoR estimates for comparators (red)

Pola-BR (DOR-IRC)

==> picture [276 x 189] intentionally omitted <==

BR (DoR-IRC)

==> picture [452 x 171] intentionally omitted <==

----- Start of picture text -----
GO29365 Vacirca et al., 2014
----- End of picture text -----

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

==> picture [452 x 222] intentionally omitted <==

----- Start of picture text -----
Ohmachi et al., 2013 Pooled naïve and MAIC-adjusted
estimates of HRs
Not available
----- End of picture text -----

Source: MAIC technical report(64)

Abbreviations: BR = bendamustine and rituximab; DoR = duration of response; HR = hazard ratio; IRC = independent review committee assessed; KM = Kaplan-Meier; pola-BR, polatuzumab vedotin with bendamustine and rituximab; TAFA+LEN = tafasitamab + lenalidomide

Response rates

The estimated ORs for ORR and complete response rate (CRR) with TAFA+LEN in the unadjusted and adjusted L-MIND populations vs. comparators are shown in Appendix D. Depth of response (ORR and CRR) in the L-MIND adjusted and unadjusted populations vs. comparators are shown in

Figure 16.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Figure 16. Depth of IRC responses for TAFA+LEN observed (green) and weighted (blue) compared with those reported for comparators (red)

MAIC vs. Pola-BR (IRC assessed)

==> picture [223 x 155] intentionally omitted <==

MAIC vs. BR (IRC assessed)

==> picture [452 x 327] intentionally omitted <==

----- Start of picture text -----
GO29365 Vacirca et al., 2014
Ohmachi et al., 2013
----- End of picture text -----

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Odds ratios for ORR for TAFA+LEN vs. BR across sources of evidence (pooled)

==> picture [186 x 232] intentionally omitted <==

Odds ratios for CR for TAFA+LEN vs. BR across sources of evidence (pooled)

==> picture [191 x 237] intentionally omitted <==

Source: MAIC technical report.(64)

Abbreviations: BR = bendamustine and rituximab; CR = complete response; HR = hazard ratio; IRC = independent review committee assessed; MAIC = matching-adjusted indirect comparison; OR = odds ratio; ORR = objective response rate; pola-BR, polatuzumab vedotin with bendamustine and rituximab; TAFA+LEN = tafasitamab + lenalidomide

Limitations of the MAIC

The shared-effect modifier assumption states that treatment effect modifiers affect all treatments in a similar way.(66) Some concerns were raised by clinical experts with respect to this assumption{Incyte Corporation, 2020 #316}. Clinical experts noted that sex could have an impact on the clearance of rituximab, and therefore, may have a differentiated effect on the efficacy of rituximab-containing regimens such as BR compared with TAFA+LEN.

Furthermore, rituximab-naïve patients were found to benefit more from R-GemOx than patients with prior rituximab exposure in the Mounier et al., 2013 study.(70) Therefore, it is possible that prior rituximab exposure could have had a differentiated impact on the efficacy of a subsequent TAFA+LEN or R-GemOx line. No population adjustment was possible on prior rituximab exposure as all patients from the L-MIND study were required to have had prior rituximab exposure. It was expected that had the Mounier et al., 2013 study(70) only included anti-CD20–experienced patients, the relative efficacy estimates would have favoured TAFA+LEN.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

In the comparison of TAFA + LEN against POLA + BR, no significant treatment benefit was observed for OS, PFS-IRC, ORR-IRC, and CRR-IRC. A numeric advantage in favour of TAFA + LEN could be observed on OS and PFS-IRC, while patients receiving POLA + BR were numerically more likely to achieve a response or a complete response compared to patients receiving TAFA + LEN, although no statistically significant treatment benefit could be estimated either. A significant treatment effect in favour of TAFA + LEN was detected on DoR-IRC both before and after population-adjustment. However this result should be interpreted with caution due to the small sample size supporting this analysis. There were some concerns about the assumption of proportional hazards for the comparison of OS and PFSIRC, and time-varying hazard ratios (HR) were estimated before and after four months from baseline. For OS and PFS-IRC, an HR initially numerically favoring POLA + BR was estimated for the first four months, followed by an HR numerically favoring TAFA + LEN. Importantly, a significant treatment effect of TAFA + LEN against POLA BR could be estimated on OS after the first four months on treatment. Although a numerical advantage in favour of TAFA + LEN could be observed after the first four months on treatment on PFS, this result was not statistically significant.

B.2.10. Adverse reactions

L-MIND

Overall extent of exposure

Overall, 427 patients received tafasitamab in the clinical study programme (as of 30 June 2019). In the primary safety analysis pool, 222 patients received tafasitamab (141 patients received tafasitamab as monotherapy and 81 as combination therapy in the L-MIND study), with an overall cumulative patient exposure of approximately 155 patient years.(61)

Long-term data from the L-MIND study (30 October 2020 data cut-off) showed that the median duration of exposure to study treatment (either TAFA+LEN or tafasitamab monotherapy) was 9.2 months (range: 0.23–53.67). The median duration of exposure to lenalidomide was xxxx weeks (range: xxxxxxxx).(57)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Treatment-emergent AEs

In the pivotal L-MIND clinical trial, treatment-emergent AEs (TEAE) of any grade occurred in all 81 patients. Neutropenia was the most common AE (all grades), occurring in 40 patients (49%). Common AEs of grade 3 or worse included thrombocytopenia, febrile neutropenia, leukopenia, anaemia and pneumonia. Most non-haematological AEs were mild (grades 1 or 2; Table 17). Of these, diarrhoea was the most common, with a median duration of 8 days (IQR: 3–24). Rashes were also common; 29 patients (n=29/81; 36%) developed rashes, most of which were grade 2 or lower. Seven patients (n=7/81; 9%) had a non-serious rash of grade 3, which was classified as allergic dermatitis in three patients and as maculopapular rash, erythematous rash, pruritus and psoriasis in one patient each. All patients recovered between two and 40 days after the event onset, but one patient with allergic dermatitis recovered with sequelae 45 days after event onset and both study drugs were discontinued in this patient.(53)

Ten (12%) of 81 patients discontinued the study during the combination therapy because of adverse events (Figure 5. L-MIND study: patient disposition (all patients enrolled)Figure 5). In total, 20 (25%) of 81 patients discontinued treatment with one or both study drugs because of adverse events during the study. LEN was discontinued in one patient with psoriasis and temporarily interrupted in two patients with allergic dermatitis. Infusion-related reactions were observed in five patients (n=5/81; 6%) and were all mild (grade 1). All occurred once during the first infusion and none required discontinuation.(53)

None of the four grade 5 AEs were AEs of special interest (AESI). No cases were suspected to be related to tafasitamab or LEN (see Section B.2.10.1. for details of deaths in the study).(53)

Table 17. TEAEs (SAS)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

The table shows treatment-emergent AEs of grade 1 or 2 occurring in at least 10% of patients and all grade 3, 4, and 5 events.

*Defined by customised Medical Dictionary for Regulatory Activities query

†One report of back pain and one report of increased blood creatinine had no toxicity grading. Source: Salles et al., 2020(53)

B.2.10.1. Serious AEs

In the primary analysis (30 November 2018 data cut-off), serious AEs (SAE) occurred in 41 patients (n=41/81; 51%). The most frequent (in two or more patients) were pneumonia (n=5/81;6%), febrile neutropenia (n=5/81;6%), pulmonary embolism (n=3/81; 4%), bronchitis (n=2/81; 2%), atrial fibrillation (n=2/81; 2%), and congestive cardiac failure (n=2/81; 2%).(53)

As of the 30 October 2020 data cut-off, 43 patients (n=43/81; 53.1%) had experienced a treatment-emergent SAE during the L-MIND study. The most frequent treatment-emergent SAEs were similar to those reported in the primary analysis.(57)

Deaths

As of the 30 November 2018 data cut-off, 30 patients died (n=30/81; 37%)—eight patients died during study treatment and 22 died post treatment. Twenty-three of the 30 deaths (77%) were related to lymphoma progression and seven (23%) were unrelated to disease progression. TEAEs leading to death occurred in four (13%) of the 30 patients: sudden death, respiratory failure, cerebrovascular accident and

© Incyte Biosciences UK (2022). All rights reserved

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

worsening of progressive multifocal leukoencephalopathy. None were considered related to the study treatment.(53)

As of the 30 October 2020 data cut-off, 42 patients had died (12 additional patients after the primary analysis; n=42/81; 51.9%). As with the primary analysis, no deaths were considered related to the study treatment.(57)

B.2.10.2. MOR208C201

TEAEs

A summary of the TEAEs reported in the DLBCL cohort and total study population of the MOR208C201 study is provided in Table 18.

Table 18. TEAEs (SAS)

Data are number of patients (%). aTEAEs according to the Medical Dictionary for Regulatory Activities preferred term (PT)

bTEAEs including PT disease progression

cTEAEs reported at grade 3 in two or more patients overall

dIn two patients, pneumonia started during the extended treatment phase (days 706 and 468, respectively), both patients recovered within two weeks. One patient developed pneumonia with cardiorespiratory failure (unrelated to tafasitamab treatment) in cycle 1 (day 23) with a fatal outcome

eNo grade 3 or grade 5 infusion-related reactions were reported.

Source: Jurczak W, et al., Ann Oncol . 2018;29:1266–72(54) Abbreviation: DLBCL = diffuse large B-cell lymphoma

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

B.2.11. Ongoing studies

There are four ongoing clinical studies investigating tafasitamab (in combination with other treatments and as monotherapy) in 1L and 2L+ R/R DLBCL:

• The pivotal L-MIND study: phase II, open-label, multicentre study characterising the safety and efficacy of tafasitamab in combination with lenalidomide in adults with R/R DLBCL(4)

• B-MIND: an open-label, phase II/III randomised, two-arm, multicentre study of tafasitamab + bendamustine vs. rituximab + bendamustine in patients with R/R DLBCL who are receiving 2L or 3L treatment and who are not candidates for highdose chemotherapy (HDC) and ASCT (thus have exhausted their therapeutic options)(71)

• An expanded access study for tafasitamab: in patients with R/R DLBCL(72)

• FIRST-MIND: a phase Ib study of tafasitamab monotherapy or TAFA+LEN, both in addition to R-CHOP, in the 1L DLBCL setting(73)

As part of the conditional marketing authorisation, Incyte has committed to complete a further clinical study with TAFA+LEN in R/R DLBCL patients not eligible for ASCT. This study is due in 2026 and could further support the evidence already presented. This could support inclusion in the Cancer Drugs Fund to support the decision making for this appraisal.

The international clinical development programme investigating tafasitama+b for the treatment of DLBCL and other cancers is summarised in Table 19.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Table 19. Clinical development programme for tafasitamab

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

Sources: NCT04134936(73); NCT02399085(4); NCT02763319(71); NCT04150328(74); NCT04300803(72); NCT01685021(75); NCT02639910(76); NCT01161511(77); NCT02005289(78); NCT01685008(79); NCT04697160(71)

Abbreviations: B-ALL = B-cell acute lymphoblastic leukaemia; BTKi = Bruton's tyrosine kinase inhibitor; CLL = chronic lymphocytic leukaemia; DLBCL = diffuse large B-cell lymphoma; NHL = non-Hodgkin lymphoma; PLL = prolymphocytic leukaemia; R/R = relapsed/refractory; SLL = small lymphocytic lymphoma; Tx = treatment; UK = United Kingdom; US = United States

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

© Incyte Biosciences UK (2022). All rights reserved

B.2.12. Innovation

Tafasitamab an Fc-enhanced mAb directed against CD19, combined with the immunomodulatory agent lenalidomide, is a novel immunological treatment combination and represents a step change in the management of R/R DLBCL. The value of this new therapeutic combination to patients with R/R DLBCL is highlighted by the Promising Innovative Medicines designation awarded by the Medicines and Healthcare products Regulatory Agency in the UK (January 2020 – PIM 2019/0012) and accelerated approval received from the US Food and Drug Administration (FDA) on the 1 July 2020. Additionally tafasitamab maintained orphan designation in R/R DLBCL after EMA and MHRA assessed that DoR could be clinically relevant and supportive of a significant benefit over Pola+BR (based on MAIC analysis).(10)

Combination treatments for patients with R/R DLBCL, including transplant-ineligible patients, commonly include re-targeting of CD20 in combination with chemotherapy, despite evidence that some B-cell malignancies, including DLBCL, lose CD20 expression after exposure to anti-CD20 therapy.(24, 36, 44, 45, 48, 80, 81)

While advances in treatment for R/R DLBCL have improved response rates in second and third line, these treatment options are not providing durable responses in majority of patients. Patients with R/R DLBCL, in particular those with advanced age, associated comorbidities, and CD20-negative transformation after prior treatment with rituximab, may be unsuitable for regimens such as ASCT, CAR-T therapy, or pola-BR.

CD19 is a transmembrane protein and signalling molecule present on B cells that is involved in B-cell development, differentiation, proliferation and signalling via enhancement of B-cell receptor signalling.(82) The protein is expressed throughout the B-cell lineage and across a wider population of B cells than CD20.(82) Tafasitamab triggers malignant B-cell death by binding to CD19 and inducing direct cytotoxicity as well as immune-mediated mechanisms, i.e., NK cell-mediated ADCC and macrophage-mediated antibody-dependent cellular phagocytosis, with ADCC as the primary contributor to the mechanism of action (MOA) for tafasitamab.(8) In

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

several in vitro studies comprising various leukaemia and lymphoma models, tafasitamab exhibited greater B-cell cytotoxic potential compared with the CD20targeting antibody rituximab.(8) LEN, an immunomodulatory agent that enhances the activity and recruitment of NK cells, has been shown to enhance NK-cell-mediated ADCC when combined with tafasitamab in vitro.(83)

An analysis of B-cell lymphoma patient biopsies revealed that CD19 expression is preserved even after CD20 is downregulated by anti-CD20 treatment.(45, 84) This allows sequencing of novel CD19 therapies with anti-CD20 treatments such as rituximab. Multiple CD19-targeting therapies have demonstrated clinical efficacy in R/R DLBCL, including adoptive cell therapies (i.e., CAR-T therapies) and antibodydrug conjugates (i.e., loncastuximab tesirine), providing clinical evidence that CD19targeting therapies deliver a treatment option for patients who progress after antiCD20 therapy.(85-91) Preclinical and clinical data also suggest that tafasitamab does not impact CD19 expression or the mechanism and viability of CAR-T therapy.(92-95) Introducing the potential for sequencing. Data on CD19 sequencing is currently immature though early research is showing promise.(54)

Finally, in contrast to currently available fixed-duration treatment options, the safety and tolerability of tafasitamab allows administration until disease progression in the majority of patients.(53, 54)

B.2.13. Interpretation of clinical-effectiveness and safety evidence

L-MIND (MOR208C203) demonstrated that TAFA+LEN, followed by tafasitamab monotherapy, resulted in deep and durable clinical responses in patients with R/R DLBCL who failed at least one prior systemic therapy (including an anti-CD20 therapy) and were not eligible for ASCT.

In the primary analysis, TAFA+LEN resulted in a best ORR of 60%, CR rate of 42.5%, PR rate of 17.5% and a median time to CR of 6.80 months. The activity of this combination was consistent across patient subgroups, including those who were refractory to prior therapies. In addition to the positive ORR result, responses were durable (median DoR was 21.7 months), particularly in patients who achieved a CR

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

(median DoR was not reached). The median PFS was 12.1 months and median OS was not reached (73.7% of patients were alive at 12 months).(53)

An updated analysis of the study was conducted with three years of follow-up for all patients (30 October 2020 data cut-off). The results of the primary analysis, including the durability of response, were confirmed with continued treatment. The median DoR was 43.9 months, median PFS 11.6 months, and median OS 33.5 months. With a median follow-up of 42.7 months for OS.(53)

At ≥35 months of follow-up, 23.5% of patients were still alive and continued to receive treatment at the data cut-off; the median OS was 33.5 months (median survival follow-up of 42.7 months.(54) For context, adults diagnosed with R/R DLBCL in a systematic review of published literature were estimated to have an agestandardised, one-year survival of 41%.(25) For patients who were refractory to 1L therapy, median OS was 6.3 months, with only 22% of patients alive at two years, in a large pooled, retrospective analysis of patients with refractory DLBCL (SCHOLAR1 study).(26) Table 20 presents the time to event endpoints split by number of prior lines.

Table 20. L-MIND extended follow-up analysis for PFS, DOR and OS by prior lines of therapy (data cut-off 30 October 2020; ≥35 months of follow-up)(54)

Abbreviations: CI = confidence interval; DoR = duration of response; NR = not reported; OS = overall survival; PFS = progression-free survival

To appropriately contextualise the data, in the absence of an RCT, two indirecttreatment comparisons using 1:1 NN matching methodology were developed (REMIND and RE-MIND2). As well, an MAIC study comparing L-MIND with an SLRbased list of prospective studies of comparators generated results that were consistent with those observed in the RE-MIND studies.

RE-MIND2 estimated the activity of TAFA+LEN in the context of the various therapies administered in routine care. Primary analysis results showed statistically

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

significant and clinically meaningful improvement of OS in the TAFA+LEN cohort vs. the cohorts of BR (31.6 vs. 9.9 months; HR=0.418 [95% CI: 0.272, 0.644]), and R- GemOx (31.6 vs. 11.0 months; HR=0.467 [95% CI: 0.305, 0.714]). All time-to-event endpoints (PFS, EFS and TTNT) supported the primary analysis results of OS and were in line with the overall results, with numerical but clinically meaningful differences, non-statistically significant differences between the TAFA+LEN cohort and the BR and R-GemOx. TAFA+LEN showed numerical improvement in these comparisons vs. pola-BR.

These results were supported by the findings of the MAIC study, which compared published literature identified by an SLR with the results from L-MIND. The MAIC analyses showed significant improvements for TAFA+LEN vs. LEN monotherapy and vs. BR. In the MAICs of TAFA+LEN vs. LEN and vs. BR, TAFA+LEN was found to significantly improve OS, PFS and CR rate, and have a numeric advantage on ORR. DoR achieved by patients was also significantly longer when receiving TAFA+LEN vs. BR or vs. Pola-BR. For OS applying a time varying hazard ratio provided a statistically significant result from four months to the end of follow-up irrespective of population adjustment for TAFA+LEN vs. Pola+BR.

Tafasitamab was well tolerated in the L-MIND study. The most frequently reported AEs were in line with the MOA for tafasitamab, mild, and managed as part of routine oncology practice.(53) Few patients had to stop tafasitamab due to AEs (12% (n/N=10/81) in the study overall).(53) The safety profile of TAFA+LEN was very similar to that of LEN alone. Comparing the commonly reported AEs to data for LEN monotherapy revealed no major difference in incidence between these two treatments.

While the safety profile of tafasitamab was typical for a B-cell targeting mAb, the incidence and severity of infusion reactions were much lower than seen with other mAbs used in the treatment of B-cell malignancies.(96, 97) Infusion-related reactions were managed with appropriate supportive/prophylactic therapy.

Prolongation of remission has been identified by international and national treatment guidelines as an important goal of therapy.(41) The results of L-MIND demonstrated

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

that TAFA+LEN can provide lasting remission and overall survival in patients with R/R DLBCL, a population that is known to be difficult to treat.

B.2.13.1. End-of-life criteria

The combination treatment of TAFA+LEN meets the NICE end-of-life criteria as summarised in Table 21.

Table 21. End-of-life criteria

Abbreviations: 1L = first line; CI = confidence interval; DLBCL = diffuse large B-cell lymphoma; FAS = full analysis set; KM = Kaplan-Meier; NHS = National Health Service; NR = not reached; OS = overall survival; R/R = relapsed/refractory

B.3. Cost-effectiveness

B.3.1. Published cost-effectiveness studies

In the economic SLR, 40 R/R DLBCL economic publications were identified, of which four were cost-effectiveness analyses assessing the cost-effectiveness of comparators included in the final scope. Table 22 presents an overview of these studies.

One study explored the cost-effectiveness of TAFA+LEN against existing treatment pathways in terms of cost per LYG using a discrete event simulation model, although the exact comparators considered were not clearly stated in the study abstract.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

The remaining three cost-effectiveness studies (Betts 2019 and Betts 2020(100), Patel 2020) compared pola-BR against BR from a US payer perspective for a transplant-ineligible R/R DLBCL population. Betts 2019 and Betts 2020(100) both adopted a partitioned survival model approach, whereas Patel 2020(101) used a Markov modelling approach.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Table 22. Overview of cost-effectiveness analysis studies

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: AE = adverse event; BR = bendamustine and rituximab; DLBCL = diffuse large B-cell lymphoma; HSCT = haematopoietic stem cell transplantation; ICER = incremental cost-effectiveness ratio; LY = life year; PD = progressive disease; pola-BR, polatuzumab vedotin with bendamustine, and rituximab; QALY = quality-adjusted life-year; R/R - relapsed or refractory; TAFA+LEN = tafasitamab + lenalidomide; US = United States; USD = United States dollar

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Further details on the economic SLR methodology and results are described in Appendix D and Appendix H.

B.3.2. Economic analysis

For the economic analysis, a de novo economic model was constructed to evaluate TAFA+LEN in the transplant-ineligible R/R DLBCL setting. To inform the main model inputs and assumptions, a review of previous health technology assessments (HTA) and relevant guidelines in transplant-ineligible R/R DLBCL was conducted, and expert opinion sought via discussions with key opinion leaders (KOL) in the UK (Table 23 and Appendix M).

B.3.2.1. Patient population

The population included in the economic evaluation were patients with R/R DLBCL ineligible for stem cell transplantation (SCT), in line with the population enrolled in the L-MIND study (Section B.2.2. ), the decision problem addressed in this submission (Section B.1.1. ), and the marketing authorisation for TAFA+LEN (Appendix C).

Patients in the model were assumed to have an average baseline age of 69.3 years, 54.5% male, a mean weight of xxxxxxx and a mean height of xxxxxxxx, based on the L-MIND population characteristics.

B.3.2.2. Model structure

An economic model was developed in Microsoft Excel[®] to assess the costeffectiveness of tafasitamab vs. relevant comparators for the treatment of patients with DLBCL who are ineligible to receive SCT in line with the licensed indication.

A partitioned survival modelling approach was selected in line with NICE Decision Support Unit (DSU) guidance. Partitioned survival models are one of the most commonly adopted modelling approaches for oncology, particularly for advanced cancer populations.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

This approach was also in line with most recent NICE technology appraisals (TA) for R/R DLBCL, including the most recent NICE Ta (TA649(5)) for pola-BR (Table 23).

Partitioned survival model approach

Figure 17 illustrates the partitioned survival model health states, which applies treatment-specific and independent OS and PFS curves for each comparator. These curves are used directly to calculate the proportion of patients in the mutually exclusive health states of pre-progression, post-progression, and death at any given time.

Figure 17. Model Diagram

==> picture [298 x 189] intentionally omitted <==

The survival partition approach does not directly calculate the transitions between health states but partitions the population into groups based on survival outcomes. At any timepoint in the model, patients falling under the PFS curve are in the preprogression health state, with the proportion of patients between the OS and PFS curves classified as having PD and the remainder above the OS curve in the death health state (Figure 18).

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Figure 18. Example survival partition approach

==> picture [452 x 234] intentionally omitted <==

----- Start of picture text -----
100%
90%
80%
48% Dead (OS)
70%
60%
50%
40% 31% PPS (PPS = OS–PFS)
30%
20%
10% 21% PFS
0%
0 50 100 150 200 250 300 350 400 450 500
28-day Cycles
Dead Progressed Progression Free
Percent of Patients
----- End of picture text -----

Abbreviations: OS = overall survival; PFS = progression-free survival; PPS = post-progression survival

B.3.2.3. Features of the economic model

Perspective

The economic analysis was performed from a UK National Health Service (NHS) and Personal and Social Services (PSS) perspective and considered only direct medical costs, including drug costs, drug administration costs (e.g., co-medications), monitoring, management of AEs, subsequent treatment costs, and disease management costs, in line with the NICE reference case.(103)

Cycle length

In line with the treatment cycle length for TAFA+LEM, a four-week cycle length was applied. This cycle length was deemed sufficiently short to accurately capture clinical outcomes and differences in treatment administrations between comparators.

Time horizon and discounting

A 45-year time horizon was used in the base case, expected to cover a lifetime horizon for patients in the target population given the median age (69.3 years) of patients in the L-MIND study. This time horizon was considered long enough to

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

capture the relevant long-term clinical and economic consequences of DLBCL for patients who are ineligible for ASCT, and was also aligned with prior NICE appraisals for R/R DLBCL therapies (Table 23).

Cost and health-related (like quality-adjusted life years [QALY]) outcomes were discounted at a rate of 3.5% in the base case in accordance with the NICE reference case.(103)

Summary of key features of the economic model vs. prior NICE R/R DLBCL TAs

The key features of the economic analysis compared to prior NICE TAs for R/R DLBCL are summarised in Table 23.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Table 23. Key features of the economic analysis

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Sources: NICE TA649 (5)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: BNF = British National Formulary; DLBCL = diffuse large B-cell lymphoma; DSU = Decision Support Unit; eMIT = electronic market information tool; HRQOL = health-related quality of life; NHS = National Health Service; NICE = National Institute for Health and Care Excellence; OS = overall survival; PD = progressive disease; PF = progression free; PFS = progression-free survival; PMBCL = primary mediastinal B-cell lymphoma; pola-BR = polatuzumab vedotin with bendamustine and rituximab; PSS = Personal and Social Services; PSSRU = Personal and Social Services Research Unit; R/R = relapsed/refractory; SCT = stem cell transplantation; SD = stable disease; SF-36 = 36-item Short Form health survey; TA = technology appraisal; TAFA=LEN = tafasitamab + lenalidomide; UK = United Kingdom

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

B.3.2.4. Intervention technology and comparators

Intervention – TAFA+LEN

The model intervention is TAFA+LEN, as described in Section B.1.2. Both tafasitamab and LEN are administered in four weekly (28 day) treatment cycles.

Tafasitamab is administered by intravenous (IV) infusion at a dose of 12 mg/kg. For the first three treatment cycles, tafasitamab is administered weekly on days 1, 8, 15 and 22 of each 28-day treatment cycle, with an additional loading dose administered on day 4 of the first treatment cycle. After the first three treatment cycles, tafasitamab is then administered on days 1 and 15 (bi-weekly) of each 28-day treatment cycle.

LEN is administered orally at a dose of 25 mg per day for days 1 to 21 of each 28day treatment cycle, up to a maximum of 12 treatment cycles.

Comparators – pola-BR, R-GemOx and BR

The comparators included in the economic analysis were pola-BR, R-GemOx and BR. These treatments were considered the most relevant comparator therapies for a R/R DLBCL population ineligible for SCT based on the R/R DLBCL patient pathway and feedback from clinical experts.{Incyte Corporation, 2020 #316} Dosing for polaBR and BR was based on NICE TA649(5), with dosing for R-GemOx based on Mounier 2013(70) and the maximum number of treatment cycles for R-GemOx based on UK lymphoma guidelines(105):

 R-GemOx:

• Rituximab 375 mg/m[2] IV on day 1 of every 15-day treatment cycle up to a maximum of six treatment cycles

• Gemcitabine 1000 mg/m[2] and oxaliplatin 100 mg/m[2] IV on day 2 of every 15-day treatment cycle up to a maximum of six treatment cycles

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

 BR

• Bendamustine 90 mg/m[2] IV on two consecutive days for each threeweek treatment cycle (days 2 and 3 of cycle 1, days 1 and 2 of cycles 2-6) up to a maximum of six total treatment cycles

• Rituximab 375 mg/m[2] IV on day 1 for each three-week treatment cycle up to a maximum of six total treatment cycles

 Pola-BR

• Polatuzumab vedotin 1.8 mg/kg IV once every three-week treatment cycle (day 2 of cycle 1, day 1 of cycles 2-6) up to a maximum of six total treatment cycles

• Bendamustine and rituximab dosing as per BR regimen

Relative efficacy estimates for comparators were obtained using results from the REMIND2 study or the MAIC (see Section B.2.9. for details).

B.3.3. Summary of base-case analysis inputs and assumptions B.3.3.1. Summary of selected base case OS and PFS methods for comparator therapies

A summary of the selected data sources and methods for OS and PFS extrapolation for the base case analysis are summarised below in Table 24. RE-MIND2 data were selected for R-GemOx and BR due to the larger sample sizes compared to the clinical trial data used for the MAIC (74 and 75 vs. 49 and 40, respectively), and the availability of patient level data to allow for more robust analyses and exploration of various parametric extrapolations. Standardised mean differences for key baseline patient characteristics showed no substantial imbalances between TAFA+LEN and R-GemOx or BR after 1:1 matching, and UK clinical experts{Incyte Corporation, 2020 #316} indicated that the RE-MIND2 parametric survival extrapolations for R-

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

GemOx and BR produced plausible estimates in relation to clinical practice. REMIND2 results for R-GemOx and BR were also robust with respect to multiple sensitivity analyses, as shown in Appendix M.1.1.

In addition, due to the poor overlap of L-MIND and Mounier 2013(70) patient populations, the population adjustment for the MAIC for TAFA+LEN vs. R-GemOx was limited. No adjustment could be made on refractoriness of patients to their prior therapy, older patients were kept in the L-MIND population while patients above 75 should not have been candidates for inclusion in the Mounier 2013(70) study, and no adjustment on the number of prior lines of therapy received by patients could be made beyond the exclusion of patients treated in the fourth-line setting or beyond in L-MIND. Therefore, the results produced by the MAIC are expected to be biased in favour of R-GemOx.

However, given the smaller sample size for the Pola-BR comparison for RE-MIND2 compared to the MAIC (39 vs. 40) and clinical expert feedback on the plausibility of the Pola-BR data from RE-MIND2, time-varying HRs from the MAIC were used instead.

Table 24. Base case modelling approaches for OS and PFS

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: BR = bendamustine + rituximab; HR = hazard ratio; MAIC = matching-adjusted indirect comparison; PFS = progression-free survival; Pola-BR = polatuzumab + bendamustine + rituximab; R-GemOx = rituximab + gemcitabine and oxaliplatin; TAFA+LEN, tafasitamab plus lenalidomide.

Base-case extrapolations for OS are shown in Figure 19, with PFS extrapolations shown in Abbreviation: KM = Kaplan-Meier

Figure 20.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Figure 19. Base case OS extrapolations

==> picture [378 x 135] intentionally omitted <==

==> picture [378 x 135] intentionally omitted <==

Abbreviation: KM = Kaplan-Meier

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Figure 20. Base case PFS extrapolations

==> picture [379 x 141] intentionally omitted <==

==> picture [379 x 141] intentionally omitted <==

Abbreviation: KM = Kaplan-Meier

The following efficacy data scenario analyses were also explored to investigate the

impact of choosing different data sources and modelling approaches:

• Alternative RE-MIND2 parametric models

  •  R-GemOx

    • Gompertz for OS

    • Generalised gamma for PFS

  •  BR

    • Generalised gamma for PFS

• Pola-BR specific scenarios:

  • Apply MAIC HRs with 11-month split

  • Apply constant MAIC HR

  • Apply adjusted parametric models from RE-MIND2 data (generalised gamma for OS, exponential for PFS)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

  - This scenario utilises TTD data from RE-MIND2 for Pola-BR to align with the data source selection for PFS 

• Applying MAIC HR estimates for all comparators (using time-varying HRs with 4-month split for Pola-BR)

  • This scenario utilises exponential models fitted to the median TTD estimates

    • from clinical trial data for R-GemOx and BR to better align with the use of clinical trial data for PFS HR estimates

B.3.3.2. Summary of base‐case analysis inputs

A summary of the key inputs for the economic analysis is shown in Table 25. Detailed parameter estimates are available in the sections references in the table or in Appendix M.

Table 25. Summary of base-case inputs

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: AE = adverse event; BR = bendamustine and rituximab; HR = hazard ratio; MAIC = matchingadjusted indirect comparison; OS = overall survival; PD = progressive disease; PFS = progression-free survival; pola-BR = polatuzumab vedotin with bendamustine and rituximab; R-GemOx = rituximab in combination with gemcitabine, oxaliplatin; TAFA+LEN = tafasitamab + lenalidomide; TTD = time to treatment discontinuation

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

B.3.3.3. Assumptions

A summary of the key modelling assumptions made are listed in Table 26.

Table 26. Key Assumptions

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 107

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

Abbreviations: BR = bendamustine and rituximab; NICE = National Institute for Health and Care Excellence; polaBR = polatuzumab vedotin with bendamustine and rituximab; R/R = relapsed/refractory; TA = technology appraisal; TAFA+LEN = tafasitamab + lenalidomide; TTD = time to treatment discontinuation

B.3.4. Measurement and valuation of health effects

B.3.4.1. Health-related quality-of-life data from clinical trials

HRQoL data were not collected in the L-MIND trial. As such, HRQoL data were sought from previous NICE appraisals and publications identified as part of the SLR.

B.3.4.2. Mapping

No mapping of HRQoL was performed for the economic analysis.

B.3.4.3. Health-related quality-of-life studies

An SLR was also performed to identify studies reporting HRQoL and health state utility data in patients with DLBCL. Further details on the SLR methodology and results are provided in Appendix G and Appendix H.

A total of 30 studies were identified in the review of HRQoL evidence. Of these, only three studies with health state utility estimates for relevant model comparators included in the final scope (Betts 2019 and Betts 2020(100), Patel 2020(101)) were identified. Health state utility values from these studies, alongside previous NICE R/R DLBCL submissions, are summarised below in Table 27.

Generally speaking, health state utility values identified from the published studies from the SLR and prior NICE R/R DLBCL studies were sourced either from older studies for aggressive non-Hodgkin lymphoma (NHL), or from the ZUMA-1 or JULIET trials which included R/R DLBCL populations receiving CAR-T therapy.

Table 27. HRQoL and utility studies in R/R DLBCL identified in the SLR and previous NICE appraisals for R/R DLBCL

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 108

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

Abbreviations: CAR-T = chimeric antigen receptor T-cell; EQ-5D, EuroQol-5 Dimension; NHL = non-Hodgkin lymphoma; PD = progressed disease; PFS = progression-free survival; R/R DLBCL = relapsed/refractory diffuse large B-cell lymphoma.

B.3.4.4. Health-related quality-of-life data used in the cost-effectiveness analysis

Utility values were applied to each health state to capture the quality of life associated with treatment and disease outcomes. Table 28 details the utilities used within the model for patients remaining progression free and alive (PFS) or with progressed disease. Although these utility data were derived from a CAR-T population which may not be generalisable to patients from the L-MIND study, base case utility estimates were sourced from the NICE appraisal for axicabtagene ciloleucel (TA559(1)), which were also applied in the NICE R/R DLBCL technology appraisal for Pola-BR (TA649(70)). In addition, two of the three UK clinical experts,

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 109

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

with whom the base case utility values were discussed with, indicated that these utility values were reasonable given their use in TA649, although one of the two clinical experts noted that progressed disease patients may have a lower health state utility as the population of patients receiving TAFA+LEN and other model comparators may be older and generally less fit than patients receiving CAR-T therapy.

Quality of life loss from subsequent CAR-T therapy was also applied in the base case analysis. A published study identified in the SLR (Lin 2019(70)) included utility lower utility estimates for CAR-T therapies for the first 2 months of therapy relative to chemoimmunotherapy. The difference in utility values between chemoimmunotherapy and tisagenlecleucel for the first 2 months of therapy (0.63 – 0.58 = 0.05) was used to generate a one-off disutility for CAR-T treatment.

Table 28. Summary of utility values for cost-effectiveness analysis

Abbreviations: PD = progressed disease; PFS = progression-free survival; SE = standard error

A second set of health state utilities from NICE TA567 (0.83 for PFS, 0.71 for PD) was also explored in scenario analysis.(21, 54)

Cure assumptions were not included in the base case analysis. However, assumption of equivalent quality of life to progression-free patients and assumption of equivalent quality of life to the general population were both explored in cure assumption-related scenario analyses.

Quality of life loss related to each adverse event (AE) was applied as a one-off QALY loss to each treatment, with disutilities and AE durations displayed in Table 29. QALY losses associated with each AE were weighted by the probability of the AE occurring for each treatment, with AE probabilities summarised in Section B.3.5.7.

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 110

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

Table 29. AE Disutilities

*Assumption - maximum Treatment disutility from TA306(35) (Pixantrone for R/R aggressive NHL) Note this assumption was used in Polatuzumab NICE submission (see p100) Abbreviations: AE = adverse event; LDH = lactate dehydrogenase

B.3.4.5. Age and sex adjustment of utilities

In order to account for differences in age and sex characteristics between the model population and reference populations for the utility values, as well as account for decreasing quality of life with increasing age, utilities were adjusted for age and sex within the model.

To adjust for age and sex, utility values applied in the model were compared against general population utility estimates for the reference age and sex characteristics of the underlying population used to derive the utility estimates. Depending on the adjustment approach selected (additive or multiplicative), an absolute utility decrement (additive) or a multiplication factor (multiplicative) were derived between the health state utility value and the age and sex-matched general population utility. This utility decrement or multiplication factor was then applied to a general population utility curve derived for the modelled population sex characteristics and age over time in order to generate a utility curve by age for each health state.

For the base case analysis, the multiplicative approach was applied under the assumption that there may be some overlap in disease symptoms or patient outcomes (such as hospitalisation events) with other age-related conditions, and in line with commentary from NICE regarding the NICE methods guide update. Reference population characteristics used to generate the disutility multiplier vs. the

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 111

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

general population for the progression-free and progressed disease health states were based on the ZUMA-1 trial (median age of 58 years, 67% male), with reference population characteristics for the utility scenario analysis based on the JULIET trial (median age of 56 years, 64.5% male). Reference population characteristics for AE disutilities were based on the original publications used to derive the AE disutilities.

General population utility was modelled according to published UK regression models from Ara/Brazier 2010(106) and Chang-Douglass 2020.(107) Both studies provide general population regression models derived from Health Survey for England (HSE) data, with Chang-Douglass 2020 updating the Ara/Brazier regression model to include additional HSE datasets for 2008, 2010-2012, 2014 and 2017. For the base case analysis, the general population regression model from ChangDouglass 2020 was applied given the larger sample size of HSE data included in the analysis, and due to the availability of uncertainty data around the regression model coefficients (which were not provided in the Ara/Brazier 2010 study).

General population and health state utility curves applied for the base case analysis are shown in Figure 21.

Figure 21. General population and age/sex adjusted health state utility curves

==> picture [437 x 132] intentionally omitted <==

==> picture [437 x 131] intentionally omitted <==

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 112

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

B.3.5. Cost and healthcare resource use identification, measurement and valuation

The economic analysis was conducted from the NHS and PSS perspective, with appropriate unit cost sources such as NHS reference costs (2019-2020), PSSRU Unit costs of Health and Social Care 2020, the British National Formulary (BNF) online (October 2021), and the drugs and pharmaceutical electronic market tool (eMIT) (September 2021) used to inform model cost inputs.

Disease- and treatment-related costs were applied to each health state and event in the model. Cost categories included: drug and administration costs applied for the duration of active treatment (determined by dosing regimen and treatment duration); routine follow-up care costs; and unplanned event costs, such as adverse events, progression, and terminal care costs.

B.3.5.1. Drug acquisition costs

Drug acquisition costs for the treatment options included in the model for induction and maintenance are shown in Table 30 and Table 31 respectively.

For TAFA+LEN and R-GemOx, patients who have not discontinued treatment by the end of induction treatment phase could move on to maintenance treatment phase. All tafasitamab patients not discontinuing prior to the end of the induction period were assumed to move on to the maintenance phase of treatment. For R-GemOx, 78% of patients remaining on treatment moved to the maintenance phase of treatment, based on Mounier 2013(70) (where 28 out of the 36 patients completing induction started the consolidation phase of treatment).

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 113

Tafasitamab Systematic Literature Review – Diffuse Large B-Cell Lymphoma | Incyte Biosciences International Sàrl

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Dose intensities are included in the model to adjust the drug costs based on the actual dosage received by the patient, and are shown in Table 30. For the L-MIND study, dose intensity for each treatment cycle for tafasitamab and lenalidomide was calculated as follows:

Dose intensity �[����� ���� ��������] ∗100 ������� ����� ����

Median dose intensity parameters for Pola-BR and BR were sourced from NICE TA649, with dose intensity estimates for Pola-BR based on the overall patient populations of the Phase Ib and Phase II trials. R-GemOx dose intensities were assumed to be 100% in the absence of available data.

For treatments with weight-based dosing, a mean weight of xxxxxx was applied based on the patient characteristics of the L-MIND trial. A body surface area (BSA) of xxxxxx was also calculated based on the mean weight (xxxxx) and mean height (xxxxxxxx) of patients in L-MIND and used to estimate drug costs for regimens with dosing based on BSA. For both weight- and BSA-based based treatments, a normal distribution around the mean weight or BSA was used to distribute the proportions of patients requiring different numbers of vials, from which a weighted average cost per dose was calculated.

No vial sharing was assumed in the base case analysis, with vial sharing for all IV based treatments explored in scenario analysis.

CONFIDENTIAL – FOR INTERNAL PURPOSES ONLY 114

Table 30. Induction Drug Costs

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

*Dose intensity describes the median intensity of dosages. Where dose intensity is not 100%, patients receive a lower dosage after a number of treatment cycles.

Ref 1: Drugs and pharmaceutical electronic market information tool (eMIT)(24) - Pharmex data for the period 01/01/20 - 31/12/20, for Pharmex products shown as Generic in the period 01/07/20 - 31/12/20. Access data: September 2021.

For gemcitabine and dexamethasone 14 different formulations were listed, therefore calculated £/mg (mg per pack) and selected 4 options with lowest £/mg Ref 2: BNF Access date: September 2021

Abbreviations: BNF = British National Formulary; BSA = body surface area; eMIT = electronic market information tool; IV = intravenous; PAS = Patient Access Scheme; PO = orally

Table 31. Maintenance Drug Costs

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

B.3.5.2. Treatment schedule

The treatment schedules for the induction phase for all comparators are summarised in Table 32.

Table 32. Induction Treatment Schedule

Abbreviations: BR = bendamustine and rituximab; CAR-T = chimeric antigen receptor T-cell; CSR = clinical study report; Pola-BR = polatuzumab + bendamustine + rituximab; R2 = lenalidomide + rituximab; R-DHAP = rituximab, dexamethasone, cytarabine, and cisplatin; R-GemOx = rituximab + gemcitabine and oxaliplatin

Treatment schedules for the maintenance phase for all comparators are summarised in Table 33. For R-GemOx, 78% of patients received consolidation treatment for cycles 5-8 in Mounier 2013(70). However, as UK guidelines for R-GemOx

recommend a maximum of 6 treatment cycles(110), 78% of patients were instead assumed to have up to 6 cycles of treatment.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Table 33. Maintenance Treatment Schedule

Abbreviations: BSA = body surface area; CSR = clinical study report; R2 = lenalidomide + rituximab; R-GemOx = rituximab + gemcitabine and oxaliplatin

B.3.5.3. Administration costs

Administration costs for IV and subcutaneous (SC) treatments included in the model are presented in Table 34, with the unit cost per resource sourced from NHS reference costs(2).

As the first and subsequent instances had different costs these were both included within the model. A radiotherapy administration unit cost is also included as some patients receive radiotherapy in the subsequent line of treatment.

Table 34. Administration Costs

Abbreviations: IV = intravenous; NHS = National Health Service; SC = subcutaneous

B.3.5.4. Concomitant medications

Table 35 details the drug dosing and cost calculation for the co-medication costs for each of the treatments.

In the L-MIND study, co-medications were given prior to tafasitamab infusion for the first three infusions. In the absence of infusion related reactions and at the discretion of the investigator, co medications were not mandated for subsequent infusions.(9)

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Otherwise, co-medications were continued for subsequent infusions. Co-medications were therefore assumed to be received all patients on TAFA+LEN for the first 4- week treatment cycle and then 0% of patients thereafter for the base case analysis. In terms of concomitant treatment with methylprednisolone, doses of between 80120mg were administered in the L-MIND study, and as such a fixed dose of 100mg was assumed for patients on TAFA+LEN.

All patients on other treatments were assumed to receive co-medications during their fixed duration treatment periods. Inclusion of co-medications was based on NICE TA649(5) for Pola-BR and BR, and El Gnaoui 2007(112) for R-GemOx.

Table 35. Co-medication Drug Dosing and Cost Calculation

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: BR = bendamustine + rituximab; Pola-BR = polatuzumab + bendamustine + rituximab; R-GemOx = rituximab + gemcitabine and oxaliplatin; NA = not applicable; Tx = treatment

Table 36. Administration Dosing for Co-medications

Abbreviations: IV = intravenous; NA = not available; PO = per oral

Table 37 displays the total co-medication costs for each of the treatments used

within the model, which were calculated using the data in Table 35 and Table 36. For TAFA+LEN, co-medication costs are only applied to the proportion of patient receiving these comedications over time (assumed to be 100% in the first cycle, then 0% of patients in subsequent cycles).

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

For IV co-medications with the same frequency of administration (such as diphenhydramine and methylprednisolone), it was assumed that these comedications would be administered simultaneously.

Table 37. Co-medication Costs

Abbreviations: BR = bendamustine and rituximab; Pola-BR = polatuzumab + bendamustine + rituximab; R- GemOx = rituximab + gemcitabine and oxaliplatin

B.3.5.5. Subsequent treatment costs

Drug costs for subsequent treatment options after progression are included in the model. These post-progression costs are a combination of possible SCT and other anti-cancer drug costs, including their administration costs.

The proportions of patients receiving different subsequent treatments upon progression on each induction treatment are listed in Table 38 and are based on the full analysis set for RE-MIND2, with the costs associated with each subsequent treatment being listed in Table 39. A 2% threshold was applied for inclusion of subsequent treatments from RE-MIND2 among any treatment arm, with the exception of CAR-T and SCT.

Table 38. Subsequent Treatment Distributions

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Abbreviations: BR = bendamustine and rituximab; CAR-T = chimeric antigen receptor T-cell; GemOx = gemcitabine and oxaliplatin; Pola-BR = polatuzumab + bendamustine + rituximab; R2, lenalidomide + rituximab; R-DHAP = rituximab, dexamethasone, cytarabine, and cisplatin; R-GemOx = rituximab + gemcitabine and oxaliplatin.

For SCT and CAR-T, any % of patients with subsequent treatment were included. However, instead of the full analysis set for RE-MIND2, subsequent CAR-T and SCT proportions were estimated using the matched RE-MIND2 patient populations to ensure balance in the underlying patient populations given the high cost of treatment with these therapies. No patients on TAFA+LEN in the matched populations received subsequent CAR-T. In the matched populations, 5.1%, 4.0% and 4.1% of patients received subsequent CAR-T following treatment with Pola-BR, BR and R-GemOx, respectively. No patients in any of the matched RE-MIND2 cohorts received subsequent SCT.

For subsequent CAR-T, it was assumed that 67% of the population received 1 cycle of Pola-BR bridging therapy between t-cell collection and t-cell re-administration based on UK clinical expert feedback, with the exception of patients receiving subsequent CAR-T therapy after Pola-BR (assuming that patients would be unlikely to receive Pola-BR bridging therapy after previously treatment with Pola-BR).

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Dosing for subsequent treatments was based on published trials, NICE technology appraisals or available treatment protocols and summary of product characteristics information.

Company evidence submission for tafasitamab with lenalidomide for treating relapsed or refractory DLBCL

©Incyte(2022). All rights reserved

Table 39. Subsequent Treatment Drug Costs