CADTH Reimbursement Review

Tafasitamab (Minjuvi)

Sponsor: Incyte Biosciences Canada Corporation

Therapeutic area: Diffuse large B-cell lymphoma (DLBCL)

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Introduction

Non-Hodgkin lymphoma (NHL) is a cancer of the immune system that encompasses more than 60 types of cancer affecting the lymphocytes.¹ In 2021, it was estimated that 11,100 of people living in Canada would be diagnosed with NHL and 2,900 of those in Canada would die from NHL that year.² Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of NHL, constituting 30% to 40% of cases in Canada.^3,4 DLBCL represents a heterogeneous group of aggressive B-cell malignancies.⁴ Some types of indolent B-cell lymphomas can transform into DLBCL (e.g., follicular lymphoma).⁵ Although the cure rate of DLBCL is high, approximately 30% to 50% of patients in Canada experience relapsed or refractory (R/R) disease after treatment with standard first-line chemotherapy with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) or a similar regimen.^4,6

Patients with R/R DLBCL have limited treatment options, ranging from supportive care to conventional salvage therapy and autologous stem cell transplant (ASCT). Eligibility for this salvage approach depends on performance status, age, and comorbidities, and eligibility for ASCT also depends on the response to salvage chemotherapy.⁴ The prognosis for patients with relapsed DLBCL who do not undergo high-dose therapy and ASCT is poor.⁴ Even for those patients who respond to salvage chemotherapy and undergo ASCT, 50% are likely to relapse following ASCT.⁴ In patients with R/R DLBCL who are not eligible for intensive therapies, there is no standard treatment approach. There are numerous chemotherapy options, but response rates are generally low and remission duration is short.⁴ Polatuzumab vedotin plus bendamustine plus rituximab (pola-BR) is an option for those living in Canada in this setting, if it is funded, according to the clinical experts consulted by CADTH.

Tafasitamab is an Fc-enhanced monoclonal antibody that targets the CD19 antigen expressed on the surface of pre-B and mature B lymphocytes and on several B-cell malignancies, including DLBCL.⁷ Tafasitamab is indicated in combination with lenalidomide for the treatment of adult patients with R/R DLBCL not otherwise specified, including DLBCL arising from low-grade lymphoma, who are not eligible for ASCT. The recommended dosage of tafasitamab is 12 mg/kg body weight administered as an IV (IV) infusion in 28-day cycles.⁷ According to the product monograph, tafasitamab should be administered with lenalidomide for up to 12 cycles. After a maximum of 12 cycles of combination therapy, patients receive tafasitamab infusions as monotherapy until disease progression or unacceptable toxicity. Tafasitamab has serious warnings and precautions in the Health Canada product monograph for infection, myelosuppression, progressive multifocal leukoencephalopathy (PML), and hepatitis B reactivation.

The objective of this review is to perform a systematic review of the beneficial and harmful effects of tafasitamab (200 mg single-use vial) in combination with lenalidomide for the treatment of adult patients with R/R DLBCL not otherwise specified, including DLBCL arising from low-grade lymphoma, who are not eligible for ASCT.

After the draft CADTH pan-Canadian Oncology Drug Review Expert Review Committee (pERC) recommendation for tafasitamab was issued in May 2022, the sponsor submitted additional post hoc analyses from the RE-MIND2 study. The results of these post hoc analyses are presented in Appendix 4. These data were not included in the initial submission to CADTH. After the CADTH recommendation was issued, the sponsor reported that the data became available only after its submission to CADTH.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups who responded to CADTH’s call for patient input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

One patient advocacy group provided input on tafasitamab for the treatment of DLBCL in adult patients. Lymphoma Canada (LC) conducted 4 anonymous online surveys. Overall, 150 DLBCL patients responded to the surveys, of which 2 (1%) indicated they had received tafasitamab therapy. Commonly reported symptoms affecting patients’ health-related quality of life (HRQoL) at diagnosis included fatigue or lack of energy, enlarged lymph nodes, drenching night sweats, unexplained weight loss, loss of appetite, influenza-like symptoms, and persistent cough. Patients also described mental and emotional problems associated with their disease and treatment that negatively affected their quality of life. Patients rated longer survival and remission than current therapies and controlling disease symptoms as the most important outcomes for a new therapy. Better HRQoL and fewer side effects compared to current therapies were also important considerations.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical experts reported that the goal of treatment in patients with R/R DLBCL who are not eligible for intensive therapies (i.e., ASCT and or chimeric antigen receptor T-cell [CAR T-cell] therapy) is to control symptoms with minimal toxicity to improve HRQoL, delay disease progression, and prolong life. The clinical experts noted that ASCT and CAR T-cell therapy both have toxicity and feasibility issues that limit broad application. Of the available options for patients who are not eligible for intensive therapies, or who have a relapse after these therapies, there is no standard of care treatment and there is no treatment that is curative (i.e., patients are treated with palliative intent). According to the clinical experts, most currently used treatment options have short durations of response, if patients respond.

The clinical experts indicated that tafasitamab in combination with lenalidomide (tafasitamab plus lenalidomide) would be an option at relapse for second-line therapy in patients who are not eligible for intensive therapy. Tafasitamab plus lenalidomide treatment could also be used in the third-line or later setting for patients who have a relapse after ASCT.

The clinical experts thought that patients who would most likely to benefit from tafasitamab plus lenalidomide are those with relapsed DLBCL, including those with underlying indolent lymphomas. The clinical experts thought that tafasitamab plus lenalidomide may be considered in patients who are not eligible for ASCT or CAR T-cell therapy, or who decline either of these treatments. The clinical experts indicated that it is not possible to identify patients who are most likely to exhibit a response to tafasitamab plus lenalidomide before treatment because there are no data on which patient or tumour characteristics are optimal for this treatment compared to other options. The clinical experts thought that patients with primary refractory DLBCL would be least suitable for treatment with tafasitamab plus lenalidomide because these patients have been excluded from the pivotal L-MIND trial. In addition, the clinical experts noted that patients who cannot come in for frequent IV infusions would not be suitable for this treatment.

The clinical experts consulted by CADTH reported that standard of care for assessing treatment response is imaging with CT (CT) or PET-CT (PET-CT) every 3 to 4 months (or sooner if there is a change in patient’s clinical status) and clinical examination and bloodwork before each treatment. The clinical experts indicated that a clinically meaningful response to treatment would include improvement in survival as well as duration of response (DOR), which would usually correlate with improvement in symptom burden. According to the clinical experts, meaningful response would include complete response (CR), partial response (PR), or stable disease with a tolerable toxicity profile.

The clinical experts noted that any disease progression should be an indication for treatment discontinuation. The clinical experts thought that recurrent infections, serious infection due to B-cell depletion, and hypogammaglobulinemia may also be considerations for discontinuation.

Clinician Group Input

Clinician input on the review of tafasitamab for the treatment of adult patients with R/R DLBCL was received from 2 groups: the Ontario Health – Cancer Care Ontario (OH-CCO) Hematology Drug Advisory Committee and a group of 4 clinicians whose submission was coordinated by LC. The clinician groups agreed that tafasitamab plus lenalidomide would be recommended in patients with DLBCL who do not respond to or relapse after first-line therapies. There were differing opinions on which patients are unsuitable for tafasitamab. The clinicians from OH-COO stated that DLBCL patients who have progressed on CAR T therapies would be least suitable for this therapy, while the LC-coordinated group maintained that there are no specific parameters that make a patient unsuitable.

Drug Program Input

The drug plans noted that primary refractory disease was an exclusion criterion in the L-MIND study and sought clarification on the definition of primary refractory disease, and whether these patients would be eligible for treatment, if the drug were reimbursed. In response, the clinical experts consulted by CADTH noted that the L-MIND study definition of primary refractory disease changed mid-study, thus complicating analysis of benefits in this high-need patient population. In the original study protocol, only patients whose disease relapsed within 3 months of a previous anti-CD20–containing regimen were defined as having primary refractory disease and excluded. After the protocol amendment, primary refractory disease was defined as disease progressing during the course of the first-line treatment as per International Working Group (IWG) response criteria (Cheson et al., [2007])⁸ and/or showing a response of less than a PR to first-line treatment or disease recurrence or progression within less than 6 months from the completion of first-line therapy. The clinical experts indicated that patients with primary refractory DLBCL are unlikely to have chemosensitive disease to subsequent therapies; these patients have an extremely poor outcome. Due to the changing definition and exclusion criterion, the clinical experts thought that it is difficult to determine whether patients with primary refractory DLBCL should be treated with tafasitamab plus lenalidomide as second-line or later therapy. The clinical experts reported that the pivotal study of pola-BR⁹ did not specifically exclude this patient population, and, therefore, pola-BR may be a better treatment option for patients with primary refractory DLBCL. However, the clinical experts also noted that there are no standard of care treatment options for these patients with high unmet need; thus, they would consider offering tafasitamab plus lenalidomide to patients with primary refractory disease.

The drug plans also asked whether the following patients would be eligible for tafasitamab: patients with a history of double- or triple-hit genetics DLBCL, patients with central nervous system (CNS) lymphoma, and patients with other histological types of lymphoma (e.g., primary mediastinal large B-cell lymphoma [PMBL] or Burkitt lymphoma). The clinical experts noted that patients with known double- or triple-hit genetics lymphoma were excluded from the L-MIND trial; thus, they indicated they would favour treatment with pola-BR in this population. The clinical experts noted that patients with a history of double- or triple-hit genetics may respond to tafasitamab plus lenalidomide, but data are not currently available to support this. The clinical experts thought that patients with CNS involvement of lymphoma, PMBL, or Burkitt lymphoma should not be treated with tafasitamab plus lenalidomide because there is no evidence to support the use of this treatment in these patients.

The drug plans inquired about whether patients who have received more than 3 prior lines of treatment, but who would otherwise fit the L-MIND trial eligibility criteria, should be eligible for tafasitamab plus lenalidomide on a time-limited basis if reimbursed. The clinical experts consulted by CADTH thought that these patients should be eligible for tafasitamab plus lenalidomide, particularly if they had no prior access to a novel therapy (i.e., CAR T-cell therapy, pola-BR). The clinical experts thought that the number of prior lines of therapy should not affect a patient’s eligibility for treatment with tafasitamab plus lenalidomide.

The drug plans also asked about the sequencing of tafasitamab plus lenalidomide therapy with pola-BR and CAR T-cell therapy. In addition, the drug plans asked how clinicians would decide when to use tafasitamab plus lenalidomide therapy versus pola-BR. The clinical experts consulted by CADTH reported that there is no evidence from prospective clinical studies to guide sequencing of tafasitamab plus lenalidomide with pola-BR and CAR T-cell therapy. The clinical experts indicated they would use tafasitamab plus lenalidomide in patients previously treated with bendamustine or polatuzumab,however they noted that there is no evidence that re-treatment with these drugs is ineffective, and, currently, they do not use these drugs routinely. The clinical experts also thought that tafasitamab plus lenalidomide may be preferred over pola-BR in patients with existing peripheral neuropathy. In addition, the clinical experts noted that the duration of pola-BR treatment is limited, whereas treatment with tafasitamab plus lenalidomide is followed by tafasitamab monotherapy until disease progression or intolerable toxicity. The clinical experts thought that some patients may prefer a treatment option with limited duration. The clinical experts noted that, in Ontario, they do not have the option to proceed with funded CAR T-cell therapy after tafasitamab is given to a patient. The clinical experts reported that pola-BR can be used for bridging for CAR T-cell therapy.

Clinical Evidence

Pivotal Studies and Protocol Selected Studies

Description of Studies

One single-arm, multi-centre, open-label, phase II study (L-MIND, N = 81) of tafasitamab plus lenalidomide in adult patients with DLBCL who had a relapse after or were refractory to 1 to 3 previous systemic regimens (with at least 1 anti-CD20 therapy), who were not candidates for high-dose chemotherapy (HDC) and subsequent ASCT, was included.^10,11 The primary objective of the L-MIND study was to determine the activity of tafasitamab plus lenalidomide in terms of objective response rate (ORR) (CR + PR) in adult patients with R/R DLBCL. Patients received IV tafasitamab (12 mg/kg) and oral lenalidomide (25 mg/day) for up to 12 cycles (28 days each), followed by tafasitamab monotherapy in patients with stable disease or better until disease progression. The primary end point was ORR by independent review committee (IRC). Other efficacy outcomes assessed included ORR by investigator assessment, overall survival (OS), progression-free survival (PFS), time to progression (TTP), event-free survival (EFS), complete response rate (CRR), DOR, time to response (TTR), and time to next treatment (TTNT). Harms outcomes were also examined. HRQoL outcomes were not reported.

In the L-MIND study, the mean age of patients was 69.3 years. Most patients were White (88.9%), had Ann Arbor stage III or IV disease (75.3%), and did not have a prior ASCT (88.9%). Overall, 54.3% of the enrolled patients were male, 55.6% had an Eastern Cooperative Oncology Group performance status (ECOG PS) of 1, 50.6% had an International Prognostic Index (IPI) score of 3 to 5, and 46.9% had disease of germinal centre B-cell like (GCB) cell origin by immunohistochemistry. Mean time since first DLBCL diagnosis was 39.6 months (standard deviation [SD] 34.8). All (100%) patients had at least 1 prior anticancer medication; 50.6% of patients had received 2 or more prior therapy lines; and 44.4% were refractory to their most recent previous therapy. The most common reasons for ASCT ineligibility were older age (46.3%) and chemorefractory status (22.5%).

Efficacy Results

Results for the key efficacy outcomes in the L-MIND study are summarized in Table 2. Three analyses were conducted based on 3 data cuts. The primary analysis had a data cut-off date of November 30, 2018.^10,11 Two additional interim analyses, which were not pre-specified in the study protocol, were conducted, with data cut-off dates of November 30, 2019, and October 30, 2020.^12-14 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||¹⁵

Table 2: Summary of Key Results From L-MIND

AE = adverse event; CI = confidence interval; CR = complete response; DCO = data cut-off; DOR = duration of response; EFS = event-free survival; FAS = full analysis set; IRC = independent review committee; NR = not reached; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; PML = progressive multifocal leukoencephalopathy; PR = partial response; SAE = serious adverse event; TTNT = time to next treatment; TTP = time to progression; TTR = time to response; WDAE = withdrawal due to adverse event.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

^cCalculated using the reverse Kaplan–Meier method, considering the censored patients as events and patients with events as censored.

^d95% CI calculated using 2-sided 95% Clopper-Pearson exact method based on binomial distribution.

^eNo patients experienced grade 3 or higher tumour lysis syndrome or cytokine release syndrome. Tumour lysis syndrome or cytokine release syndrome events of any grade were not reported.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Overall Survival

At the primary analysis, median OS was not reached (NR) (95% confidence interval [CI], 18.3 to NR) with a median follow-up time of 19.6 (95% CI, 15.3 to 21.9) months. As of the most recent analysis, the median OS was 33.5 months (95% CI, 18.3 to NR), with a median follow-up time of 42.7 (||||||||||||||||||||||||) months.

Progression-Free Survival

At the primary analysis, median PFS by IRC was 12.1 months (95% CI, 5.7 to NR), with a median follow-up time of 17.3 (95% CI, 11.5 to 21.2) months. As of the most recent analysis, median PFS by IRC was 11.6 months (95% CI, 6.3 to 45.7), with a median follow-up time of 33.9 (||||||||||||||||||||||||) months.

Time to Progression

At the primary analysis, median TTP was 16.2 months (95% CI, 7.4 to NR). TTP was not analyzed at the subsequent interim analyses.

Event-Free Survival

At the primary analysis, median EFS was 9.1 (95% CI, 5.3 to 21.0) months, with a median follow-up time of 19.7 (95% CI, 14.3 to 22.0) months. EFS was not reported at the subsequent interim analyses.

Objective Response Rate

ORR by IRC was the primary end point in L-MIND. At the primary analysis, the ORR by IRC was 60.0% (95% CI, 48.4 to 70.8). The best objective response for patients was CR for 34/80 (42.5%) patients and PR for 14/80 (17.5%) patients. As of the most recent interim analysis, ORR by IRC was 57.5% (95% CI, 45.9 to 68.5). Thirty-two (40.0%) patients had CR and 14 (17.5%) patients had PR.

Duration of Response

At the primary analysis, median DOR by IRC was 21.7 (95% CI, 21.7 to NR) months. Median DOR by IRC in patients with PR was 4.4 months (95% CI, 2.0 to 9.1) and NR (95% CI, 21.7 to NR) in patients with CR. As of the most recent interim analysis, median DOR by IRC was 43.9 (95% CI, 26.1 to NR) months. Median DOR by IRC in patients with PR was 5.6 (95% CI, 2.2 to NR) months compared to NR (95% CI, 43.9 to NR) in patients with CR.

Time to Response

At the primary analysis, median TTR (CR or PR) based on IRC evaluation was 2.0 months (range 1.7 to 16.8 months). At the second analysis, median TTR based on IRC evaluation was 2.0 months (range 1.7 to 16.8 months). TTR was not reported at the most recent interim analysis.

Time to Next Treatment

At the primary analysis, median TTNT was 15.4 (95% CI, 7.6 to NR) months. At the second analysis, median TTNT was 12.5 (95% CI, 7.6 to 24.7) months. TTNT was not reported at the most recent interim analysis.

Health-Related Quality of Life

HRQoL outcomes were not reported in L-MIND.

Harms Results

Harms data from the L-MIND study safety analysis set (N = 81) as of the most recent analysis (October 30, 2020, data cut-off) are summarized in Table 2. As of both the primary analysis and most recent analysis, the median duration of exposure to the study treatment (tafasitamab plus lenalidomide) was 9.2 months.

Adverse Events

All 81 (100%) patients enrolled in L-MIND experienced at least 1 treatment-emergent adverse event (AE). The most common AEs were neutropenia (50.6%), anemia (37.0%), diarrhea (35.8%), thrombocytopenia (30.9%), and cough (27.2%).

Serious Adverse Events

Overall, 53.1% of patients enrolled in L-MIND experienced at least 1 serious adverse event (SAE). The most common SAEs were pneumonia (n = 7, 8.6%), febrile neutropenia (n = 5, 6.2%), and pulmonary embolism (n = 3, 3.7%). Other SAEs reported in more than 1 patient included bronchitis, lower respiratory tract infection, atrial fibrillation, and congestive cardiac failure (n = 2, 2.5% each).

Withdrawals Due to Adverse Events

Overall, 20 (24.7%) patients permanently discontinued treatment with 1 or both study drugs due to AEs: 8 (9.9%) patients discontinued lenalidomide only, 2 (2.5%) discontinued tafasitamab only, and 10 (12.3%) discontinued both study drugs. The only AE that led to permanent discontinuation of study drug in more than 1 patient was neutropenia (n = 3, 3.7%).

Mortality

In total, 42 (51.9%) patients enrolled in L-MIND had died as of the October 30, 2020, data cut-off date. The cause of death was reported to be related to disease progression for 31 (38.3%) patients and unrelated to disease progression in 10 (12.3%) patients.

Notable Harms

Overall, 72.8% of patients enrolled in L-MIND experienced an infection. The most common types of infections were bronchitis (16.0%), pneumonia (12.3%), urinary tract infection (12.3%), and respiratory tract infection (11.1%).

Regarding myelosuppression, 50.6% of patients experienced neutropenia, 37.0% experienced anemia, 30.9% experienced thrombocytopenia, 14.8% experienced leukopenia, 12.3% experienced febrile neutropenia, and 7.4% experienced lymphopenia.

One (1.2%) patient developed worsening PML. Two (2.5%) patients experienced hepatitis B virus (HBV) reactivation. Five (6.2%) patients experienced an infusion-related reaction. No patients experienced grade 3 or higher tumour lysis syndrome or cytokine release syndrome. Tumour lysis syndrome or cytokine release syndrome events of any grade were not reported.

Critical Appraisal

For the primary end point and multiple secondary end points (i.e., PFS, EFS, DOR, TTR), an IRC was appropriately used. Furthermore, there was generally good agreement between the IRC and investigator-assessed outcomes. The CADTH review team and clinical experts evaluated the eligibility criteria and analysis populations as appropriate.

L-MIND is an open-label, single-arm study. There is no direct evidence comparing tafasitamab plus lenalidomide to a control arm. Furthermore, no statistical testing was performed because the L-MIND study was not designed to test hypotheses. Data were analyzed descriptively. Due to these limitations of the study design, the CADTH review team could draw no definitive conclusions from the L-MIND study regarding the efficacy and safety of tafasitamab plus lenalidomide relative to relevant comparators. The open-label design can increase the risk of performance and detection bias, particularly for outcomes that are subjective in measurement and interpretation (e.g., response, AEs). Objective outcomes, such as OS time and mortality, are unlikely to be affected by performance or detection bias. The potential for detection bias was minimized by using IRC assessment for key study outcomes, such as ORR, DOR, and PFS. The time-to-event analyses were appropriate, but causality cannot be inferred in a single-arm trial without a comparator.

Multiple protocol amendments were implemented while the L-MIND study was being conducted, which included changes to the study eligibility criteria. There was a high rate of protocol deviations, which creates uncertainty in the data because protocol deviations could have affected the internal validity of the study. In addition, protocol deviations related to eligibility criteria may have caused selection bias, although the direction of this potential bias is unknown. The L-MIND study consisted of patients with R/R DLBCL diagnosed as per local pathologic analysis. However, central pathologic analysis concluded that approximately 10% of these patients had non-DLBCL histology or alternative diagnoses. This causes selection bias. The clinical experts consulted by CADTH indicated that inclusion of these patients may have confounded results, particularly for OS and PFS, although sensitivity analyses conducted using patients with a DLBCL diagnosis confirmed by central pathologic analysis were generally consistent with the main analysis. Overall, the L-MIND study was a phase II trial that enrolled 80 patients in the full analysis set (FAS). The clinical experts consulted by CADTH indicated that it may not be possible to extrapolate efficacy results from this small sample of patients to the general population of patients with R/R DLBCL in Canada.

The L-MIND trial was an international, multi-centre study, but there were no sites in Canada. The treatment regimen used in the L-MIND trial aligns with Health Canada–recommended dosage of tafasitamab plus lenalidomide. The clinical experts indicated that the baseline characteristics of patients enrolled in L-MIND were generally representative of the R/R DLBCL patient population in Canada, although they noted that the L-MIND study patients would represent the most fit patients in this population. The L-MIND study excluded some groups of patients in the R/R DLBCL patient population, specifically, patients with known double- or triple-hit genetics DLBCL at study entry, which limits the generalizability of results to this patient population. In addition, primary refractory DLBCL was an exclusion criterion in L-MIND. However, the definition changed during the study, complicating the interpretation of the generalizability of study results to this group of patients. Multiple protocol amendments related to the eligibility criteria were implemented during the L-MIND study, which increased the generalizability of results because the trial population was more representative of the general patient population of those living in Canada, according to the clinical experts consulted by CADTH. HRQoL outcomes, which are important to patients, were not reported, which is a key gap in the evidence.

Indirect Comparisons

Description of Studies

Three sponsor-submitted indirect treatment comparisons (ITCs) were included in this review: 2 retrospective observational studies (RE-MIND^16,17 and RE-MIND2¹⁸) that were used as external cohorts for indirect comparison with patients enrolled in the L-MIND trial, using estimated propensity score (ePS)-based nearest neighbour (NN) 1:1 matching methodology; and 1 ITC that used unanchored matching-adjusted indirect comparisons (MAICs). These ITCs were used to inform the pharmacoeconomic models.

RE-MIND^16,17 was designed to characterize the effectiveness of lenalidomide monotherapy in the treatment of R/R DLBCL patients not eligible for HDC followed by ASCT. The primary end point was ORR. Other end points assessed included OS, CRR, DOR, PFS, TTNT, and EFS. Data from the L-MIND study used in RE-MIND were from the November 30, 2018, data cut-off (primary analysis).

RE-MIND2¹⁸ was designed to characterize the effectiveness of systemically administered therapies in the treatment of R/R DLBCL patients (second, third, or fourth line). Eligible systemic therapies included regimens administered in routine clinical care according to National Comprehensive Cancer Network (NCCN) or European Society for Medical Oncology (ESMO) guidelines^19,20 for patients who were not eligible for ASCT. This study included the following treatment cohorts: systemic therapies pooled, bendamustine plus rituximab (BR), rituximab plus gemcitabine plus oxaliplatin (R-GemOx), CAR T-cell therapy, and pola-BR. The primary end point was OS. Other end points assessed included ORR, CRR, DOR, PFS, TTNT, EFS, treatment discontinuation due to AEs, and duration of treatment exposure. Data from the L-MIND study used in RE-MIND2 were from the November 30, 2019, data cut-off. The pre-specified main analysis was conducted for systemic therapies pooled, BR, and R-GemOx. Pre-specified analyses could not be conducted for pola-BR and CAR T-cell therapy due to insufficient patient numbers. Thus, only post hoc, exploratory analyses were conducted.

Unanchored MAICs²¹ of tafasitamab plus lenalidomide in the L-MIND study versus comparator therapies using prospective studies were conducted. In total, |||||||| prospective studies reporting data for ||||||||||||||||||||||||||||||||||||, pola-BR, BR, and R-GemOx were selected for the MAICs against tafasitamab plus lenalidomide. End points assessed included OS, PFS, DOR, ORR, and CRR. Data were used from the L-MIND study analysis with the October 30, 2020, data cut-off.

Efficacy Results

In RE-MIND, ORR was 67.1% (95% CI, 55.4 to 77.5) in the tafasitamab plus lenalidomide cohort compared to 34.2% (95% CI, 23.7 to 46.0) in the lenalidomide-monotherapy cohort (odds ratio [OR] = 3.885; 95% CI, 1.900 to 8.142; P < 0.0001). Median OS was NR (95% CI, 15.5 to NR) in the tafasitamab plus lenalidomide cohort and 9.4 (95% CI, 5.1 to 20.0) months in the lenalidomide-monotherapy cohort (hazard ratio [HR] = 0.499; 95% CI, 0.317 to 0.785; P = 0.0026). Median PFS was 12.1 (95% CI, 5.9 to NR) months in the tafasitamab plus lenalidomide cohort and 4.0 (95% CI, 3.1 to 7.4) months in the lenalidomide-monotherapy cohort (HR = 0.463; 95% CI, 0.307 to 0.698; P = 0.0002). Median DOR was 20.5 (95% CI, 12.3 to NR) months in the tafasitamab plus lenalidomide cohort and 6.6 (95% CI, 4.1 to 17.2) months in the lenalidomide-monotherapy cohort (P < 0.0001).

In RE-MIND2, patients in the tafasitamab plus lenalidomide cohort showed an improvement in OS compared to the cohorts of systemic therapies pooled (HR = 0.553; 95% CI, 0.358 to 0.855; P = 0.0076), BR (HR = 0.418; 95% CI, 0.272 to 0.644; P < 0.0001), and R-GemOx (HR = 0.467; 95% CI, 0.305 to 0.714; P = 0.0004). An improvement was also observed for PFS in the tafasitamab plus lenalidomide cohort compared with the cohorts of systemic therapies pooled (HR = 0.424; 95% CI, 0.278 to 0.647; P < 0.0001), BR (HR = 0.527; 95% CI, 0.344 to 0.809; P = 0.0033), and R-GemOx (HR = 0.433; 95% CI, 0.288 to 0.653; P < 0.0001). The ORR was higher in the tafasitamab plus lenalidomide cohort compared to the cohorts of systemic therapies pooled (|||||||||||||||||||||||||||||||||||||||||||||||| P = 0.0323) and R-GemOx (||||||||||||||||||||||||||||| P = 0.0076). There was no difference in ORR in the tafasitamab plus lenalidomide cohort compared to the BR cohort (|||||||||||||||||||||||||||||||||||||||||||||||| P = 0.1810).

In the MAICs, ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. For the comparisons of tafasitamab plus lenalidomide to pola-BR, no differences were observed for OS, PFS by IRC, ORR, and CRR. Overall, the results of some of the comparisons to BR favoured tafasitamab plus lenalidomide, whereas others indicated no difference. Last, in the MAIC of tafasitamab plus lenalidomide versus R-GemOx, results indicated no difference between tafasitamab plus lenalidomide and R-GemOx for all outcomes assessed.

Harms Results

In RE-MIND2, 8 patients (14.5%, 14.5%, and 15.1% in the analysis sets for comparison to systemic therapies pooled, BR, and R-GemOx, respectively) discontinued due to AEs in the tafasitamab plus lenalidomide cohort. In the cohorts of systemic therapies pooled, BR, and R-GemOx, 5 (6.8%), 2 (2.8%), and 4 (5.4%) patients, respectively, had AEs leading to permanent discontinuation of treatment. The types of AEs leading to treatment discontinuation were not reported. The median duration of exposure in the tafasitamab plus lenalidomide cohort was longer (approximately 10 months) than in the cohorts of systemic therapies pooled (2.4 months), BR (3.2 months), and R-GemOx (2.9 months). Harms outcomes were not reported in RE-MIND or the MAICs.

Critical Appraisal

The RE-MIND and RE-MIND2 studies implemented multiple measures to minimize bias. However, important sources of heterogeneity between the L-MIND cohort and observational cohorts could not be accounted for with the methods used. Although the eligibility criteria for enrolment in RE-MIND and RE-MIND2 were based on the eligibility criteria used in the L-MIND trial, differences related to the retrospective nature of the RE-MIND and RE-MIND2 studies studies were noted. Comparison of data from a prospective, interventional trial to retrospective, observational studies using real-world data may be problematic, as a number of notable differences in data collection, outcomes, and assessments were noted (e.g., tumour assessment frequency, imaging modalities, and criteria used to assess response). Most important, unmeasured confounding factors not accounted for in the matching may have had an effect on results. The RE-MIND and RE-MIND2 studies used 9 covariates for matching in their main analyses (age, Ann Arbor stage, refractoriness to last therapy line, number of previous lines of therapy, history of primary refractoriness, prior ASCT, neutropenia, anemia, and elevated LDH). Other known confounders were not accounted for in the matching (e.g., ECOG PS, IPI score, cell of origin) in the main analyses. As a result of these limitations, there is substantial risk of bias in the RE-MIND and RE-MIND2 study results.

There are also limitations to the external validity of the RE-MIND and RE-MIND2 studies. Lenalidomide monotherapy is not used as a treatment for R/R DLBCL in Canada, according to the clinical experts. RE-MIND-2 included relevant comparators, but the clinical experts consulted by CADTH also indicated that R-GemOx and BR are not very commonly used to treat patients with R/R DLBCL in Canada. The clinical experts indicated that pola-BR would be the most relevant comparator, although it is not yet funded. The clinical experts noted that the relevance of CAR T-cell therapy as a comparator for tafasitamab plus lenalidomide in patients who are not eligible for ASCT was debatable. The clinical experts considered CAR T-cell therapy an intensive therapy and thus more comparable to ASCT. The clinical experts indicated that they would not consider using tafasitamab plus lenalidomide in patients who were eligible for CAR T-cell therapy. There are also concerns about whether the systemic therapies pooled cohort adequately reflects current contemporary practice and therapies in Canada.

Although the methods used to conduct the unanchored MAICs followed technical guidance,²² the analyses have limitations that affect internal and external validity. Most important, not all known effect modifiers and prognostic factors identified by the authors could be adjusted for in the analyses, due to the lack of available data. The quality of most of the comparator studies was low. Furthermore, multiple sources of heterogeneity (e.g., study design, eligibility criteria, study end point definitions, timing of tumour assessments) were identified and could not be accounted for in the analyses conducted. Given these issues, there is substantial concern about the risk of bias in the MAIC results. There are also limitations to the external validity of some of the comparators (i.e., lenalidomide monotherapy, BR, and R-GemOx), as previously described. In addition, results may be generalizable only to patients similar to those enrolled in the comparator studies, which may not represent patients typically seen in practice in Canada.

Other Relevant Evidence

No long-term extension studies or additional relevant studies were included in the sponsor’s submission to CADTH.

Conclusions

One phase II, single-arm, open-label trial (L-MIND) of tafasitamab plus lenalidomide in patients with R/R DLBCL was included in the systematic review conducted by CADTH. The L-MIND trial data were analyzed descriptively; no statistical hypotheses were tested. According to the clinical experts consulted by CADTH, the results suggested that tafasitamab plus lenalidomide therapy is clinically effective in this patient population and that there may be a beneficial effect of tafasitamab plus lenalidomide on OS, PFS, ORR, DOR, and other efficacy outcomes. However, there is significant uncertainty because it is a phase II trial and because of its open-label single-arm design and small sample size. Due to the absence of a comparator arm and statistical testing, the CADTH review team could draw no definitive conclusions regarding the efficacy of tafasitamab plus lenalidomide based on the L-MIND trial. HRQoL outcomes were not reported in the L-MIND trial, which represents an important gap in the evidence. All study patients reported treatment-emergent AEs, the most common which was neutropenia, and more than half reported SAEs. The most frequently reported SAEs were pneumonia, febrile neutropenia, and pulmonary embolism. The most common cause of death was disease progression.

No direct evidence on the relative efficacy and safety of tafasitamab plus lenalidomide versus other therapies was identified. Results from the ITCs submitted by the sponsor suggested that tafasitamab plus lenalidomide therapy may be associated with an improvement in clinical outcomes (e.g., ORR, CRR, OS, PFS, EFS, DOR, and TTNT) compared to lenalidomide monotherapy, systemic therapies pooled, BR, R-GemOx, pola-BR, and CAR T-cell therapies. However, the ITCs were associated with substantial risk of bias due to important limitations, including methodological limitations, heterogeneity, matching based on a limited number of variables, and small sample sizes. In view of the uncertainty in the ITC results, the CADTH review team could draw no conclusions on the efficacy of tafasitamab plus lenalidomide compared to other therapies used to treat patients with R/R DLBCL who are ineligible for ASCT. Harms outcomes were assessed in 1 ITC (RE-MIND2). The ITC results showed that a numerically greater proportion of patients treated with tafasitamab plus lenalidomide may discontinue treatment due to AEs than patients treated with systemic therapies pooled, BR, and R-GemOx. However, there were limitations associated with the data (i.e., differences in study design, data collection methods, and duration of exposure to treatment). The potential benefits and safety of tafasitamab plus lenalidomide compared with other therapies remain uncertain.

Introduction

Disease Background

NHL is a cancer of the immune system that encompasses more than 60 types of cancer affecting the lymphocytes.¹ In 2021, it was estimated that 11,100 of those living in Canada would be diagnosed with NHL and 2,900 of those in Canada would die from NHL that year.² The signs and symptoms of NHL vary depending on the type of NHL, where it starts in the body, and how advanced it is.²³ Common symptoms of NHL include swollen or enlarged lymph nodes in the neck, armpit, or groin; rash or itchy skin on the chest, stomach, and back; and unexplained fatigue.²³ Other systemic symptoms can include unexplained persistent fever, drenching night sweats, and unexplained weight loss.²³ DLBCL is the most common subtype of NHL, constituting 30% to 40% of cases in Canada.^3,4 DLBCL represents a heterogeneous group of aggressive B-cell malignancies.⁴ Some types of indolent B-cell lymphomas (e.g., follicular lymphoma) can transform into DLBCL.⁵

Although the cure rate of DLBCL is high, approximately 30% to 50% of patients in Canada experience R/R disease after treatment with standard first-line chemotherapy with R-CHOP or a similar regimen.^4,6 According to the clinical experts, patients with R/R DLBCL typically have disease confirmed on repeat biopsy. There are multiple factors associated with a worse prognosis in this patient population (e.g., GCB cell of origin, double- or triple-hit genetics, older age, greater ECOG PS score).^21,24,25

Standards of Therapy

Patients with R/R DLBCL have limited treatment options, ranging from supportive care to conventional salvage therapy and ASCT, with the choice of therapy depending on age and comorbidities. According to the clinical experts consulted by CADTH, patients with R/R disease after first-line therapy are assessed for eligibility for intensive therapy. Intensive therapies include HDC followed by ASCT and CAR T-cell treatment. According to the clinical experts consulted by CADTH, approximately half of R/R DLBCL patients are not eligible for intensive therapy due to age or comorbidities, and these patients are treated with palliative intent.

For eligible patients, the standard treatment approach for patients R/R DLBCL is an intensive salvage chemotherapy regimen followed by ASCT.^4,6 However, eligibility for this salvage approach largely depends on performance status, age, and comorbidities, and eligibility for ASCT also depends on the response to salvage chemotherapy.⁴ Among patients who progress following frontline treatment, only 30% to 40% respond to salvage chemotherapy and proceed with ASCT.⁴ The prognosis for patients with relapsed DLBCL who do not undergo HDC followed by ASCT is poor.⁴ Even among those patients who respond to salvage chemotherapy and undergo ASCT, 50% are likely to relapse following ASCT.⁴ If patients do not show chemosensitivity to salvage therapy (and thus are not transplant-eligible), or relapse following ASCT, they may be eligible for CAR T-cell therapy.⁴ Patients who are not candidates for intensive therapies due to comorbidities, age, or functional status are usually treated with palliative intent. According to the clinical experts consulted by CADTH, the goals of treatment for patients with R/R DLBCL who are not eligible for intensive therapy are to control symptoms with minimal toxicity, improve quality of life, delay disease progression, and prolong life.

In patients with R/R DLBCL who are not eligible for intensive therapies, there is no standard treatment approach. There are numerous chemotherapy options, but response rates are generally low and remission duration is short.⁴ Pola-BR would be an option for those living in Canada, in this setting, if it were funded. Pola-BR may be used as a stand-alone treatment or may provide a bridge to future consolidative therapies, including ASCT or CAR T-cell therapy.⁴ Chemotherapy options for fit patients can include rituximab, ifosfamide, carboplatin, and etoposide; rituximab, oxaliplatin, cytosine arabinoside, and dexamethasone; R-GemOx; and rituximab, dexamethasone, cisplatin, and cytarabine. Palliative strategies can also be employed.

Drug

Tafasitamab is an Fc-enhanced monoclonal antibody that targets the CD19 antigen expressed on the surface of pre-B and mature B lymphocytes and on several B-cell malignancies, including DLBCL.⁷ Upon binding to CD19, tafasitamab mediates B-cell lysis through apoptosis and immune effector mechanisms. In in vitro laboratory studies conducted in DLBCL tumour cell lines, tafasitamab plus lenalidomide was associated with greater cytotoxicity than when cells were treated with either drug alone.

Tafasitamab is indicated in combination with lenalidomide for the treatment of adult patients with R/R DLBCL not otherwise specified, including DLBCL arising from low-grade lymphoma, who are not eligible for ASCT. Key characteristics of tafasitamab and lenalidomide are summarized in Table 3. The sponsor’s reimbursement request is per the Health Canada indication. Tafasitamab has not been previously reviewed by CADTH.

Tafasitamab received a Notice of Compliance with Conditions on August 19, 2021, pending the results of trials to verify its clinical benefit. According to the Health Canada product monograph, authorization was based on ORR, CRR, and durability of response from a single-arm clinical study; an improvement in PFS or OS has not been established.⁷ According to the Letter of Undertaking,²⁴ the planned confirmatory study is a phase III, randomized, double-blind, placebo-controlled trial comparing the efficacy and safety of tafasitamab plus lenalidomide in addition to R-CHOP versus R-CHOP in previously untreated, high-intermediate, and patients at high risk patients with newly diagnosed DLBCL (the frontMIND study).^15,24 ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

The recommended dosage of tafasitamab is 12 mg/kg body weight administered as an IV infusion in 28-day cycles according to the following schedule: during cycle 1, infusions are administered on days 1, 4, 8, 15, and 22; in cycles 2 and 3, infusions are administered on days 1, 8, 15, and 22 of each cycle; and, in cycle 4 and subsequent cycles, infusions are administered on days 1 and 15 of each cycle.⁷ According to the product monograph, tafasitamab should be administered with lenalidomide for up to 12 cycles. Patients take lenalidomide capsules orally at the recommended starting dosage of 25 mg daily on days 1 to 21 of each cycle. The starting and subsequent dosages of lenalidomide may be adjusted as needed. After a maximum of 12 cycles of combination therapy, lenalidomide treatment is stopped, and patients receive tafasitamab infusions as monotherapy until disease progression or unacceptable toxicity.

Table 3: Key Characteristics of Tafasitamab and Lenalidomide

ASCT = autologous stem cell transplant; CD = cluster of differentiation; DLBCL = diffuse large B-cell lymphoma; IL = interleukin; IV = IV; MM = multiple myeloma; PML = progressive multifocal leukoencephalopathy; R/R = relapsed or refractory; TNF = tumour necrosis factor.

^aHealth Canada–approved indication.

Source: Tafasitamab Product Monograph,⁷ Lenalidomide Product Monograph.²⁶

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The full original patient input(s) received by CADTH have been included in the stakeholder section at the end of this report.

One patient advocacy group provided input on tafasitamab for the treatment of R/R DLBCL in adult patients. LC conducted 4 anonymous online surveys for DLBCL patients from April 2018 to December 2021. The most recent survey, conducted from October 2021 to December 2021, included patients receiving tafasitamab. Overall, 150 DLBCL patients responded to the surveys, of whom 2 patients (1%) indicated they had received tafasitamab therapy. The following is a summary of key input from the perspective of the patient group.

• Patients described the negative impact of DLBCL on their day-to-day life. Commonly reported symptoms affecting patients’ HRQoL at diagnosis included fatigue or lack of energy, enlarged lymph nodes, drenching night sweats, unexplained weight loss, loss of appetite, influenza-like symptoms, and persistent cough. Patients also described mental and emotional problems associated with their disease and treatment that negatively affected their HRQoL, including fear of disease recurrence, memory loss, anxiety/worry, problems concentrating, difficulty sleeping, loss of sexual desire, stress of diagnosis, and depression.

• Patients had received at least 1 line of treatment or were undergoing first-line treatment (with the most commonly reported first-line treatment being R-CHOP). Patients reported that treatment had a significant impact on their ability to work, travel, and participate in daily activities, as nearly all patients reported at least 1 side effect. Fatigue, nausea and/or vomiting, “chemo-brain,” and hair loss were reported as the most difficult side effects to tolerate. Treatment-related side effects reported by 2 patients who received tafasitamab, including neutropenia, rash or itching, diarrhea, and nausea, were short-term and did not affect the patients’ HRQoL. The majority of DLBCL symptoms resolved after patients received tafasitamab, including enlarged lymph nodes, abnormal blood cell counts (platelets, red blood cells, white blood cells), and weight loss and poor appetite.

• Patients rated longer survival and remission than current therapies and control of disease symptoms as the most important outcomes for a new therapy. Better HRQoL and fewer side effects compared to current therapies were also important considerations. Some patients (n = 48/114, 42%) indicated that they would tolerate potential side effects of a new treatment if symptoms were short-term and would choose a new treatment with known side effects (potentially serious) if their doctors recommended it as the best option. For other patients (n = 54/114, 47%), choosing a new treatment would depend on the type of side effects, exact length of time they would experience side effects, and whether side effects would outweigh treatment benefit and result in a long-term outcome or cure.

The patient group also stated that patients who are ineligible for ASCT have limited treatment options and indicated a significant unmet need in this population.

Clinician Input

Input From Clinical Experts Consulted by CADTH

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

Unmet Needs

The clinical experts reported that the goal of treatment in patients with R/R DLBCL who are not eligible for intensive therapies (i.e., ASCT and CAR T-cell therapy) is to control symptoms with minimal toxicity to improve HRQoL, delay disease progression, and prolong life. The clinical experts noted that ASCT and CAR T-cell therapy both have toxicity and feasibility issues that limit broad application. Of the available options for patients who are not eligible for ASCT or CAR T-cell therapy, or relapse after these therapies, there is no treatment that is curative. According to the clinical experts, most currently used treatment options have short durations of response, if patients respond at all. The clinical experts indicated that, at relapse of DLBCL, non-intensive treatments are usually prescribed to older adult patients (i.e., ≥ 70 years old), and this population often has comorbidities and may be frail. Thus, tolerability of treatment is important.

Place in Therapy

According to the clinical experts consulted by CADTH, initial therapy for patients with DLBCL who do not have double-hit genetics disease is R-CHOP for 3 to 6 cycles, with or without radiation. Patients with double-hit genetics DLBCL are often treated with more aggressive regimens (e.g., dose-adjusted rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and hydroxydaunorubicin). The clinical experts reported that approximately half of patients with R/R DLBCL are not eligible for intensive therapy (i.e., multi-drug chemotherapy as second-line followed by ASCT or CAR T-cell therapy) due to their age or comorbidities. If patients are not eligible for intensive therapy, there is no standard of care treatment. According to the clinical experts, if ASCT or CAR T-cell treatment are not options, or if the patient relapses following these treatments, the patient would be treated with palliative intent. These treatments aim to improve symptoms, control disease, and prolong life in some cases. Treatment options include steroids, radiotherapy, BR, and non-curative chemotherapy regimens (e.g., rituximab, gemcitabine, cisplatin, and dexamethasone; or rituximab, dexamethasone, cisplatin, and cytarabine). The clinical experts indicated that pola-BR could also be used in this patient population, if it were funded. The clinical experts reported that there is no current standard of care regimen in this population, and oncologists tend to use a regimen that they feel the patient can tolerate based on their prior experience with chemotherapy, the pace of disease progression, and the comorbidities they have.

The clinical experts indicated that tafasitamab plus lenalidomide would be an option at relapse for second-line therapy in patients who are not eligible for intensive therapy. Tafasitamab plus lenalidomide treatment could also be used in the third-line or later setting for people who relapse after ASCT. If both this treatment and pola-BR were available, the clinical experts indicated that they were uncertain how to sequence them, based on the available evidence. The clinical experts thought that patients with peripheral neuropathy or those at higher risk of myelosuppression might do better with tafasitamab plus lenalidomide than pola-BR. The clinical experts also noted that patients with underlying indolent lymphoma were excluded from the pivotal randomized trial evaluating pola-BR,⁹ so this may be another reason to consider tafasitamab plus lenalidomide. The clinical experts indicated that, if tafasitamab were reimbursed, it would add a potential option for older patients with comorbidities, rather than shifting the treatment paradigm.

In addition, the clinical experts noted that exposure to tafasitamab or any other CD19 antibody would make a patient ineligible for CD19 CAR T-cell therapy in Ontario. Therefore, they would not use tafasitamab plus lenalidomide as a bridging option to CAR T-cell therapy or ASCT. As a result, the clinical experts thought that tafasitamab plus lenalidomide would be an option only for those patients who fail all intensive treatment regimens or for whom an intensive regimen is not an option.

Patient Population

The clinical experts indicated that patients with R/R DLBCL should be diagnosed using biopsy whenever feasible. At times, patients have PR to initial therapy and then progression at this same site. In these cases, repeat biopsy is not feasible, and diagnosis can be made based on clinical reasoning that an alternative diagnosis is unlikely and treatment should be started for probable relapse, according to the clinical experts consulted by CADTH. The clinical experts agreed that both symptomatic and asymptomatic patients with DLBCL require treatment, because it is an aggressive type of lymphoma. The clinical experts indicated that a hematologist or medical oncologist assesses eligibility for treatments at relapse, and, if the patient is not eligible for intensive therapy, then tafasitamab plus lenalidomide would be considered.

The clinical experts consulted by CADTH thought that patients most likely to benefit from tafasitamab plus lenalidomide are those with a relapse of DLBCL, including those with underlying indolent lymphomas. The clinical experts thought that this treatment would be considered in patients who are not eligible for ASCT or CAR T-cell therapy or decline either of these treatments. The clinical experts indicated that it is impossible to identify patients who are most likely to exhibit a response to tafasitamab plus lenalidomide before treatment because there are no data on which patient or tumour characteristics are optimal for this treatment compared to other options.

The clinical experts thought that patients with primary refractory DLBCL would be least suitable for treatment with tafasitamab plus lenalidomide, since these patients have not been studied in trials. In addition, the clinical experts noted that patients who cannot come in for frequent IV infusions, want a time-limited treatment, or for whom intensive therapy might be considered in the future would not be suitable for this treatment.

Assessing Response to Treatment

The clinical experts consulted by CADTH reported that standard of care for assessing treatment response is imaging with CT or PET-CT every 3 to 4 months, with clinical examination and bloodwork before each treatment. Treatment response may be assessed with imaging sooner than every 3 to 4 months if there is a change in clinical status. The clinical experts indicated that a clinically meaningful response to treatment would include improvement in survival as well as DOR, which would usually correlate with improvement in symptom burden. According to the clinical experts, meaningful response would include CR, PR, or stable disease with a tolerable toxicity profile.

Discontinuing Treatment

The clinical experts noted that any disease progression should be an indication for treatment discontinuation. The clinical experts thought that recurrent infections, serious infection due to B-cell depletion, and hypogammaglobulinemia may also be considerations for discontinuation. One clinical expert also thought that inability to tolerate lenalidomide or tafasitamab may be a consideration for discontinuation because there are unclear benefits of monotherapy for either. By contrast, the second clinical expert indicated that they would not stop single-drug treatment if it provided a clinical benefit for the patient.

Prescribing Conditions

The clinical experts thought that treatment with tafasitamab could be carried out in any setting that can monitor for infusion-related reactions and has protocols and processes to deal with a hypersensitivity reaction. The clinical experts noted that this is an outpatient regimen, and standard supportive measures, such as those available for rituximab or other monoclonal antibodies, would be required. The patients would typically have standard pre-medications and management of hypersensitivity with monitoring.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The full original clinician group input(s) received by CADTH have been included in the stakeholder section at the end of this report.

Clinician input on the review of tafasitamab for the treatment of adult patients with R/R DLBCL was received from 2 groups: the OH-CCO Hematology Drug Advisory Committee and a group of 4 clinicians whose submission was coordinated by LC.

Both groups agreed that there are poor and limited treatment options for patients with R/R DLBCL. The clinician groups indicated that CAR T-cell therapies and other combinations of palliative drug regimens (e.g., pola-BR) are options for these patients. Patients who are unfit or ineligible for intensive therapies (i.e., ASCT or CAR T-cell therapy) or who have failed prior cellular therapy have the greatest unmet need for treatment, according to the clinician groups. Both groups agreed that the most important goals in treatment are prolonging survival and reducing disease symptoms. The LC-coordinated group added that ensuring a reasonable safety profile, manageable toxicities, ease of administration, and improvement in HRQoL, as well as helping the patients gain their independence and reducing burden on caregivers, are also important.

Both groups agreed that tafasitamab plus lenalidomide would be recommended after first-line therapies. The clinician groups indicated that patients with R/R DLBCL not otherwise specified, including DLBCL arising from low-grade lymphoma, who are not eligible for ASCT would be most suitable for this therapy. The LC-coordinated group added that tafasitamab plus lenalidomide would be routinely offered in patients ineligible for CAR T-cell therapy. The OH-CCO clinicians suggested that tafasitamab plus lenalidomide would be an additional option for second-line treatment, while the LC-coordinated group stated that this therapy would be used in the third-line setting or beyond. The LC-coordinated group also mentioned that most clinicians would proceed with intensive salvage therapy first and reserve tafasitamab plus lenalidomide for after failure of salvage therapy.

The LC-coordinated group added that, for patients who fail tafasitamab plus lenalidomide treatment, pola-BR is an option. Additionally, oral prednisone, etoposide, procarbazine, and cyclophosphamide are a palliative oral chemotherapy, and participation in clinical trials is an option as well. Both groups reported that receiving tafasitamab plus lenalidomide could affect patients’ eligibility for subsequent CAR T-cell therapy.

Both groups agreed that patients would be identified by their primary treating physician. They also agreed that tafasitamab plus lenalidomide can be administered on an outpatient basis in certified centres. There were 2 differing opinions on which patients are unsuitable for tafasitamab. The clinicians from OH-COO stated that patients with DLBCL who have progressed on CAR T-cell therapies would be least suitable for this therapy, while the LC-coordinated group maintained that there are no specific parameters that would make a patient unsuitable. The LC-coordinated group thought that the patient should be well enough to tolerate the frequent outpatient visits required for this therapy. Both groups agreed that there are no predictors to identify which patients will exhibit response.

Both groups provided input on how response to treatment is assessed. The group from OH-COO suggested that treatment response should be assessed every 3 months in the first year of treatment and then every 6 months after the first year of treatment. The LC-coordinated group added that clinical assessment before each cycle of treatment is the standard of practice, which may include a review of symptoms, a physical examination, and assessment of lymphadenopathies, organomegaly, and extranodal involvement. The LC-coordinated group also provided insights on additional practices related to assessing response. Imaging studies with CT and/or PET scan, if clinically indicated, are part of the assessment after cycle 4 and after cycle 12 of treatment (i.e., 4 months and 12 months after starting the treatment).

Both groups agreed that improved symptoms, prolonged remission, and improved survival would be considered a clinically meaningful response to treatment. The LC-coordinated group added that clinically meaningful responses would include resolution of all lymphoma-related symptoms, improvement in functional status and HRQoL indicators, and return to normal activities. Both groups agreed that confirmation of disease progression would constitute treatment failure and would prompt treatment discontinuation. The LC-coordinated group suggested that severe toxicities (e.g., grade 3 or higher) should result in temporary discontinuation until improved, and that therapy should be discontinued if the toxicity is unacceptable to either the patient or the physician provider.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

BL = Burkitt lymphoma; CAR T-cell therapy = chimeric antigen receptor T-cell therapy; CD = cluster of differentiation; CEOP = cyclophosphamide, epirubicin, vincristine, and prednisone; CHOP = cyclophosphamide, doxorubicin, vincristine, and prednisone; CNS = central nervous system; DLBCL = diffuse large B-cell lymphoma; IV = IV; IWG = International Working Group; pERC = pCODR Expert Review Committee; PFS = progression-free survival; pola-BR = polatuzumab vedotin plus bendamustine plus rituximab; PMBL = primary mediastinal (thymic) large B-cell; PR = partial response; R-CEOP = rituximab, cyclophosphamide, epirubicin, vincristine, and prednisone; R-CHOP = rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone.

Clinical Evidence

The clinical evidence included in the review of tafasitamab is presented in 3 sections. The first section, the Systematic Review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review. No additional relevant studies were included in the sponsor’s submission to CADTH.

After the draft CADTH pCODR Expert Review Committee (pERC) recommendation for tafasitamab was issued in May 2022, the sponsor submitted additional post hoc analyses from the RE-MIND2 study, which provided indirect evidence on tafasitamab. The results of these post hoc analyses are presented in Appendix 4. These data were not included in the initial submission to CADTH. After the CADTH recommendation was issued, the sponsor reported that the data became available only after it had made its submission to CADTH.

Systematic Review (Pivotal and Protocol Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of tafasitamab (200 mg single-use vial) in combination with lenalidomide for the treatment of adult patients with R/R DLBCL not otherwise specified, including DLBCL arising from low-grade lymphoma, who are not eligible for ASCT.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 5. Outcomes included in the CADTH review protocol reflect outcomes considered important to patients, clinicians, and drug plans.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist (https://www.cadth.ca/resources/finding-evidence/press).²⁹

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946–) via Ovid and Embase (1974–) via Ovid. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Minjuvi/tafasitamab. Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on December 15, 2021. Regular alerts updated the search until the meeting of the CADTH pERC on April 13, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool for Searching Health-Related Grey Literature checklist (https://www.cadth.ca/grey-matters).³⁰ Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

These searches were supplemented by reviewing bibliographies of key papers.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

In addition, a focused literature search for NMAs dealing with DLBCL was run in MEDLINE All (1946–) on December 15, 2021. No date or language limits were applied.

Table 5: Inclusion Criteria for the Systematic Review

AE = adverse events; ASCT = autologous stem cell transplantation; BR = bendamustine plus rituximab; CAR T-cell therapy = chimeric antigen receptor T-cell therapy; CRR = complete response rate; DLBCL = diffuse large B-cell lymphoma; DOR = duration of response; ECOG PS = Eastern Cooperative Oncology Group performance status; EFS = event-free survival; EORTC-QLQ-C30 = European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire; GDP = gemcitabine plus dexamethasone plus cisplatin; HRQoL = health-related quality of life; IPI = International Prognostic Index; IV = IV; MEP = methotrexate plus etoposide plus cisplatin; ORR = objective response rate; OS = overall survival; PEP-C = prednisone plus etoposide plus procarbazine plus cyclophosphamide; PFS = progression-free survival; PML = progressive multifocal leukoencephalopathy; pola-BR = polatuzumab vedotin plus bendamustine plus rituximab; PRR = partial response rate; RCT = randomized controlled trial; R/R = relapsed and/or refractory; R-GemOx = rituximab plus gemcitabine plus oxaliplatin; SAE = serious adverse events; TTNT = time to next treatment; TTR = time to response; WDAE = withdrawal due to adverse events.

^aDouble-hit status is defined as genetic lesions in BCL2 and MYC; triple-hit status is defined as genetic lesions in BCL2, MYC, and BCL6.

Findings From the Literature

Eight reports^10-14,31,32 of 1 unique study (L-MIND) were identified from the literature for inclusion in the systematic review (Figure 1). The included study is summarized in Table 6. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 6: Details of Included Study

AE = adverse event; ALT = alanine transaminase; AP = alkaline phosphatase; ASCT = autologous stem cell transplant; AST = aspartate aminotransferase; 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 PS = Eastern Cooperative Oncology Group performance status; EFS = event-free survival; IRC = independent review committee; IV = IV; IWG = International Working Group; NK = natural killer; NOS = not otherwise specified; OL = open-label; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; PK = pharmacokinetic; PMBL = primary mediastinal (thymic) large B-cell; PR = partial response; THRLBCL = T-cell or histiocyte-rich large B-cell lymphoma; TTNT = time to next treatment; TTP = time to progression; TTR = time to response; ULN = upper limit of normal.

Note: Four additional reports were included from the sponsor’s submission to CADTH.^10,12,14,31

^aAdditionally, patients with evidence of histological transformation to DLBCL from an earlier diagnosis of low-grade lymphoma (e.g., an indolent pathology such as follicular lymphoma, marginal zone lymphoma, chronic lymphocytic leukemia) into DLBCL with a subsequent DLBCL relapse were also eligible.

^bRelapsed disease was defined as the appearance of any new lesions or an increase in size by 50% or more of previously involved sites, according to the 2007 IWG response criteria, after the most recent systemic therapy. Refractory disease was defined as disease progression per IWG response criteria, showing less than a PR or disease recurrence or progression in less than 6 months from the completion of first-line therapy, or showing less than a PR to the most recently administered systemic therapy.

^cThe lesion must have a greatest transverse diameter of 1.5 cm or more and greatest perpendicular diameter of 1.0 cm or more at baseline. The lesion must be positive on PET scan.

^dAs the combination of tafasitamab plus lenalidomide had not previously been evaluated in a clinical study, an evaluation of safety data was conducted after the first 6 patients had been accrued. These patients were enrolled sequentially with a 48-hour lag period between enrolments. Patient accrual was held until all 6 patients had been followed for 1 treatment cycle. Once the sixth patient in the cohort had completed the cycle 1 day 22 visit, the safety review panel performed an evaluation based on the number and type of AEs during the first cycle as well as laboratory values (biochemistry and hematology).

Source: Salles et al. (2020),¹¹ L-MIND Clinical Study Report,¹⁰ L-MIND Statistical Analysis Plan.²⁴

Description of Studies

L-MIND (N = 81) was a single-arm, multi-centre, open-label, phase II study of tafasitamab plus lenalidomide in adult patients with DLBCL who had relapsed after, or were refractory to, 1 to 3 previous systemic regimens (with at least 1 anti-CD20 therapy), who were not candidates for HDC and subsequent ASCT.^10,11 The primary objective of the L-MIND study was to determine the efficacy of tafasitamab plus lenalidomide in terms of ORR (CR + PR) in adult patients with R/R DLBCL. Patients received IV tafasitamab (12 mg/kg) and oral lenalidomide (25 mg/day) for up to 12 cycles (28 days each), followed by tafasitamab monotherapy in patients with stable disease or better until disease progression. Patients were enrolled from 35 sites in 10 countries (Europe and US).

The design of the L-MIND study is summarized in Figure 2. The L-MIND study consisted of 2 parts, which were performed sequentially. The first part was the safety run-in, during which 6 patients were enrolled in the study and completed the first cycle of study treatment. Once the sixth patient in the cohort completed the visit in cycle 1 day 22, a safety review panel performed a clinical safety review. The sponsor opened the second part of the study for enrolment based on the outcome of this interim safety evaluation. The study duration per patient could be up to approximately 5 years, including periods of screening (up to 28 days from the time the patient signed the informed consent form), the treatment period (maximum 12 cycles for tafasitamab plus lenalidomide followed by tafasitamab monotherapy thereafter), and the survival follow-up phase. The survival follow-up phase began with the end-of-treatment (EOT) visit. Patients had an onsite visit for a safety evaluation 30-days after the last administration of study treatment. All patients who discontinued for any reason (other than withdrawal of consent or death) were contacted every 90 days by telephone from the date of the 30-day safety follow-up visit. The survival follow-up covered a duration of up to 5 years from cycle 1 day 1 or until withdrawal of consent or death, whichever came first.

Figure 2: L-MIND Study Design

EOT = end of treatment; FU = follow-up; LEN = lenalidomide; MOR00208 = tafasitamab; R/R DLBCL = relapsed or refractory diffuse large B-cell lymphoma; SD = stable disease.

Source: L-MIND Clinical Study Protocol.²⁴

Three analyses based on 3 data cuts were conducted for L-MIND. The primary analysis for L-MIND had a data cut-off date of November 30, 2018, and outcomes were reported after a median follow-up of 13.2 months.^10,11 A second efficacy analysis of key outcomes was conducted based on a data cut-off date of November 30, 2019; this analysis was used in the Health Canada approval for tafasitamab plus lenalidomide.¹² A third analysis of key outcomes was undertaken with a data cut-off of October 30, 2020, to report long-term outcomes after 35 months or more of follow-up for ORR.^13,14 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The analyses with the November 30, 2019, and October 30, 2020, data cut-off dates were not pre-specified in the study protocol. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Populations

Inclusion and Exclusion Criteria

Adult patients (> 18 years of age) were eligible for the L-MIND study if they had histologically confirmed DLBCL (including transformed indolent lymphoma with a subsequent DLBCL relapse); had disease that relapsed after, or was refractory to, at least 1, but no more than 3, systemic regimens (with at least 1 anti-CD20 therapy); and were not candidates for HDC and subsequent ASCT. Patients were also required to have adequate organ function, an ECOG PS of 0 to 2, and measurable disease at baseline. Patients ineligible for ASCT included patients who were older than 70 years, who had organ dysfunction or comorbidities, who had failed previous ASCT, who did not respond to salvage therapy, who had refused ASCT, or who were unable to receive ASCT because of an inability to successfully collect peripheral blood stem cells. Patients were excluded if they had primary refractory DLBCL, double-hit or triple-hit genetics DLBCL (i.e., simultaneous detection of MYC with BCL2 and/or BCL6 translocation), ASCT within the previous 3 months, previous allogenic stem cell transplantation, or CNS lymphoma involvement. Additional details regarding inclusion and exclusion criteria in the L-MIND study are provided in Table 7.

Baseline Characteristics

Baseline characteristics of all patients enrolled in the L-MIND study are presented in Table 7. The mean age of enrolled patients was 69.3 years (SD 9.5 years). Most patients were White (88.9%), had Ann Arbor stage III or IV disease (75.3%), and did not have a prior ASCT (88.9%). Overall, 54.3% of enrolled patients were male, 55.6% had an ECOG PS of 1, 50.6% had an IPI score of 3 to 5, and 46.9% had disease of GCB cell of origin by immunohistochemistry, whereas 8.6% had disease of GCB cell of origin by gene-expression profiling. Mean time since first DLBCL diagnosis was 39.6 months (SD 34.8 months). All (100%) patients had 1 or more prior anticancer medication, 50.6% of patients had received 2 or more prior therapy lines, and 44.4% were refractory to their most recent previous therapy. The most common reasons for ASCT ineligibility were older age (46.3%) and chemorefractory status (22.5%).

Table 7: Summary of Baseline Characteristics in L-MIND – All Enrolled Patients

ABC = activated B-cell; ASCT = autologous stem cell transplantation; DLBCL = diffuse large B-cell lymphoma; ECOG PS = Eastern Cooperative Oncology Group performance status; GCB = germinal centre B-cell like; HDT = high-dose chemotherapy; IPI = International Prognostic Index; SD = standard deviation.

^aChemorefractory patients included patients who failed to achieve PR or CR with salvage therapy or who underwent stem cell transplant before enrolment.

^bAll patients who were not chemorefractory and who had comorbidities, defined as (1) diffusion lung capacity for carbon monoxide less than 50% by pulmonary function test, (2) left ventricular ejection fraction less than 50% by multiple gated acquisition echocardiogram, or (3) other organ dysfunction or comorbidities precluding the use of HDT or ASCT on the basis of unacceptable risk of treatment.

^cAll patients who were not chemorefractory, had no comorbidities as defined in [b], but were too old (age > 70 years).

Source: L-MIND Clinical Study Report.¹⁰

Interventions

All patients had received R-CHOP or equivalent chemoimmunotherapy before study entry.²⁴ Study treatment in L-MIND consisted of tafasitamab plus lenalidomide for up to 12 cycles (28 days each), followed by tafasitamab monotherapy in patients with stable disease or better until disease progression.

Tafasitamab (12 mg/kg, IV) was administered over approximately 2 hours. During cycles 1 to 3, tafasitamab infusions were administered weekly on days 1, 8, 15, and 22, with an additional loading dose administered on day 4 of cycle 1. From cycle 4 forward, tafasitamab infusions were administered every 14 days, on days 1 and 15 of each cycle.

Pre-medications as prophylaxis for infusion-related reactions were administered 0.5 to 2 hours before tafasitamab. Pre-medications included antipyretics, histamine (H₁ and H₂) receptor blockers, glucocorticoids, and meperidine. Pre-medication for patients who did not experience any infusion-related reactions to tafasitamab during the first 3 infusions was optional for subsequent infusions at the discretion of the investigator. Otherwise, pre-medication was continued for all subsequent tafasitamab administrations.

Tafasitamab was interrupted for any grade 2 to 4 infusion-related reactions or other toxicities, and patients discontinued tafasitamab if they experienced a grade 4 infusion-related reaction. Tafasitamab was administered until disease progression, unacceptable toxicity, or discontinuation for any other reason. Continuing tafasitamab despite disease progression was at the discretion of the investigator.

Lenalidomide was self-administered on days 1 to 21 of each 28-day cycle for a maximum of 12 cycles. The starting dosage was 25 mg orally daily. A stepwise dosage reduction of lenalidomide was done in cases of toxicities, with a decrease by 5 mg per day in each step, only once per cycle, without re-escalation. If the lenalidomide dosage was reduced during the previous cycle, then that reduced dosage level was continued in the new cycle. There was no more than 1 dosage reduction from 1 cycle to the next. Once a patient’s lenalidomide dose had been reduced, no dosage re-escalation was permitted.

Concomitant Medications

Patients could receive concomitant medications for the treatment of symptoms, AEs, and intercurrent illnesses. Medications to treat concomitant diseases (e.g., diabetes, hypertension, bronchial asthma, chronic obstructive pulmonary disease) were allowed. Patients could also receive therapy to mitigate side effects of the study medication, as clinically indicated, as well as best supportive care as per institutional guidelines. This included antiemetics, antidiarrheals, anticholinergics, antispasmodics, antipyretics, antihistamines, analgesics, antibiotics, and other medications intended to treat symptoms. Growth factors could be prescribed during the treatment and follow-up periods at the investigator’s discretion.

Patients were not permitted to receive a CD20-targeted therapy, chemotherapy, radiotherapy, investigational anticancer therapy, or other lymphoma-specific therapy within 14 days before cycle 1 day 1 of the study. In addition, no radiotherapy (including limited field radiotherapy) was permitted after the baseline PET-CT scan for initial disease assessment. Other than the study drugs, patients did not receive any other DLBCL-specific therapy during the study treatment period. After disease progression had been recorded, additional antineoplastic therapies were permitted at the discretion of the investigator and in accordance with the local guidelines.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trial included in this review is provided in Table 8. These end points are further summarized in this section.

Efficacy and Tumour Imaging Assessments

Disease response was assessed according to the revised response criteria based on the guidelines of the IWG reported by Cheson et al. (2007).⁸ In the L-MIND study, disease and response assessments as well as tumour measurements were performed. All response assessments were evaluated locally and centrally. Response assessment made by the central radiology and clinical review at cycle 12 day 28 and/or at the EOT visit, as applicable, were considered for the main efficacy analysis.

Table 8: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

CRR = complete response rate; DOR = duration of response; EFS = event-free survival; HRQoL = health-related quality of life; IRC = independent review committee; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; TTNT = time to next treatment; TTP = time to progression; TTR = time to response.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3,¹⁴ and sponsor’s submission to CADTH.²⁴

A local assessment of efficacy or disease response was performed and recorded at cycle 3 day 1, cycle 5 day 1, cycle 7 day 1, and cycle 10 day 1, in addition to central review at the end of cycle 12 (for decisions concerning the additional cycle 13 through cycle 24 tafasitamab treatment) and at the EOT visit (to determine whether disease had progressed). During the additional treatment phase, disease was assessed as follows:

• Cycle 13 until cycle 24: The first assessment during this phase was carried out approximately 3 months and 2 days after the cycle 12 day 28 PET-CT assessment, if required. Thus, this assessment was not needed at cycle 13 day 1, and an interval longer than 3 months was acceptable for the cycle 16 day 1 assessment. Thereafter, the assessment was repeated approximately every 3 months plus or minus 2 days from the previous scan.

• Cycle 25 onward: appoximately once every year after the previous scan.

Initial disease and disease response assessments for the primary end point were made centrally by PET-CT (CT with IV contrast) at screening, at day 28 (plus or minus 4 days) of cycle 12, and at the EOT visit, if a patient was withdrawn from treatment before the end of cycle 12 for reasons other than progressive disease (PD). PET-CT scanning during the additional treatment phase (cycle 13 onward) with tafasitamab was performed at the discretion of the investigator and was not performed more than approximately once per year.

MRI was permitted in lieu of CT, and PET-MRI in lieu of PET-CT, for patients with contraindications to the administration of contrast drugs, or due to other medical reasons, at the same time points as CT, or in addition to CT, at the discretion of the investigator.

End Point Definitions

OS was defined as the time from the first administration of any study drug until the date of death from any cause. Patients who were alive as of the data cut-off date or dropped out early were censored at the date of last contact.

PFS was defined as the time elapsed between the first administration of any study drug and lymphoma progression or death from any cause, whichever occurred first. The date of progression was the first date for which PD was assessed as the objective response. The tumour assessments were derived according to the IWG treatment response criteria for malignant lymphoma by an IRC. If a patient was alive and progression-free at the data cut-off date, the patient was censored.

TTP was defined as the time from the first administration of any study drug until documented DLBCL progression or death due to lymphoma.

EFS was defined as the time from the date of the first administration of any study drug to the date of tumour progression, first initiation of a new non-study antineoplastic therapy, or death from any cause, whichever comes first.

ORR was defined as the proportion of patients with a CR or PR up until disease progression, based on central radiological/clinical evaluations by the IRC. Objective response was defined as CR or PR after DLBCL evaluation using the IWG treatment response criteria for malignant lymphoma. The best ORR was the primary end point, considering all response assessments recorded until the cut-off for the primary analysis.

CRR was defined as the proportion of patients with a CR.

DOR was defined as the time interval between the initial time point of tumour response (CR or PR, whichever status was recorded first) and the first date that recurrence of PD or death was documented. If a patient was alive and progression-free at the data cut-off date, the patient was censored.

TTR was defined as the time from the first administration of any study drug to the first documented response (CR or PR).

TTNT was defined as the time from the first administration of any study drug to the institution of next antineoplastic therapy (for any reason, including disease progression, treatment toxicity, and patient preference) or death from any cause, whichever came first.

The incidence and severity of AEs were a secondary outcome in the L-MIND study. MedDRA-coded AEs were used to show the incidence of all AEs by system organ class, preferred term, relationship to treatment, severity, and seriousness.

Statistical Analysis

For the L-MIND study, data were analyzed descriptively. No formal statistical hypothesis testing was performed. For continuous variables, the number of non-missing observations, mean, SD, minimum and maximum values, and quartiles (first quartile, median, third quartile) were presented. For categorical variables, the number of non-missing observations, the number of missing observations, the relevant percentage of the analysis population, and frequencies were reported. If not defined otherwise, the percentage denominator was the number of patients with non-missing information. For subcategories, the relative frequencies were calculated on the basis of the patients in the respective category.

Sample Size Determination and Power Calculation

The L-MIND study planned to enrol approximately 80 patients with R/R DLBCL. The sample size was determined using various possible monotherapy and combination effect ORR rates and power assumptions. To determine the sample size, it was assumed that the combination treatment could improve the ORR from a value of 20% (with monotherapy) to 35%. Applying an exact binomial test with a 2-sided significance level of 5% and a power of 85%, the estimated sample size was 73 patients. According to this scenario, an observed ORR of 32% would lead to a statistically significant study outcome. Assuming a dropout rate of 10%, a total sample size of approximately 80 patients was estimated.

Primary End Point

ORR was defined as the proportion of responders whose best response at any time throughout treatment or follow-up was CR or PR. The number and percentage of patients classified as having a best overall response of CR or PR as well as 95% CIs calculated using the Clopper-Pearson exact method were presented.

For the main analysis, the denominator for calculating the ORR was the total number of patients in the FAS population. The tumour assessments were performed by the IRC. Patients with no post-baseline assessment of response were included as nonresponders. Patients with a best response of “not evaluable” were summarized by the reason that their status was unknown. Concordance between IRC assessment and investigator assessment was also analyzed.

Subgroup Analyses

ORR was reported for the following subgroups identified in the CADTH systematic review protocol: refractory to last therapy (yes versus no), primary refractoriness (yes versus no), prior ASCT (yes versus no), number of prior treatment lines, reason for ASCT ineligibility (chemorefractory versus older age or comorbidities versus other), cell of origin, and IPI score. Subgroup analyses were exploratory. All subgroups except for IPI score were pre-specified in the study protocol. All subgroup analyses were conducted using the FAS, based on both IRC and investigator assessment.

Sensitivity Analyses

Sensitivity analyses were conducted based on local investigator assessment, using a per-protocol set (i.e., patients without any post-baseline assessment of DLBCL response or patients who had major protocol deviations were excluded) and excluding the patients with no post-baseline assessment of response or with all post-baseline assessments categorized as “unknown.” The number (and percentage) of patients was descriptively tabulated by categories of individual best outcome in tumour response assessments (CR, PR, stable disease, PD, not evaluable).

Secondary End Points

For time-to-event end points, Kaplan–Meier estimates of the first quartile, median, and third quartile, along with their 95% CIs, were reported. The corresponding Kaplan–Meier survival curves were presented. The OS and PFS probabilities at specific time points (1 month, 3 months, 6 months, 12 months, 18 months, 24 months, 36 months, 48 months) and the associated 95% CIs (Greenwood formula) were summarized.

An AE summary table was used to present the number of events, number of patients, and percentage of patients with treatment-emergent AEs, SAEs, and treatment-emergent AEs that led to study discontinuation.

Censoring

For OS, patients who were alive or dropped out early were censored at the date of last contact. For PFS and DOR, if a patient was alive and progression-free at the data cut-off date, the patient was censored. Patients who had not experienced a lymphoma progression or death from any cause were censored at the last available tumour assessment. The same held true for patients who were lost to follow-up.

Subgroup Analyses

Subgroup analyses were exploratory. PFS, DOR, and OS were reported for the following subgroups identified in the CADTH systematic review protocol: refractory to last therapy (yes versus no), primary refractoriness (yes versus no), prior ASCT (yes versus no), number of prior treatment lines, reason for ASCT ineligibility (chemorefractory versus older age or comorbidities versus other), cell of origin, and IPI score. All subgroup analyses were conducted using the FAS based on both IRC and investigator assessment. Kaplan–Meier curves were provided for PFS, DOR, and OS.

Sensitivity Analyses

Sensitivity analyses for the secondary end points were generally performed in the same manner as for the primary end point. For PFS, an additional sensitivity analysis with patients having more than 1 missed visit, but having an available death date, were included in the time-to-event analysis and considered to have a PFS event (FAS for both IRC and investigator response assessment). For DOR, an additional sensitivity analysis was performed—with DOR stratified by best overall response—and duration of CR was analyzed.

Planned Analyses

A safety run-in analysis was planned for after the first 6 patients had been accrued. At this time, a safety review panel consisting of the sponsor, a representative of the participating investigators, and 2 independent expert clinical hematologists performed a clinical safety review based on the number and type of AEs during the first cycle and on laboratory values (biochemistry and hematology). After discussion of these data, the panel would consider the lenalidomide dosage tolerated or in need of reduction. The sponsor opened the second part of the study for enrolment following the outcome of this discussion.

The primary analysis was conducted after all patients had undergone at least 1 post-baseline response assessment at the earliest. The sponsor decided when to perform the primary analysis (approximately 12 months after the last patient has been enrolled). The primary analysis corresponds to the analysis with a data cut-off date of November 30, 2018.

The final analysis is planned when all patients have completed their EOT visit after database lock.

Protocol Amendments and Deviations

In the original L-MIND study protocol, patients were required to have a histologically confirmed diagnosis of DLBCL (not otherwise specified) according to the Revised European American Lymphoma or WHO classification. Patients with NHL other than DLBCL with classical histology (e.g., including patients with DLBCL transformed from indolent lymphomas) were excluded. In addition, patients who had relapsed within 3 months of prior CD20-targeted therapy were excluded. For the first 6 months (6 cycles) of the study, each cycle consisted of a tafasitamab infusion on day 1, day 8, day 15, and day 22 of the cycle. Additionally, a loading dose was administered on day 4 of cycle 1. Thereafter, tafasitamab was administered biweekly (every 14 days), with infusions on days 1 and 15 of each 28-day cycle. Tafasitamab could be given for up to 24 months in total. A summary of key protocol amendments implemented in the L-MIND study is presented in Table 9. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Table 9: Summary of Key Protocol Amendments in L-MIND

Anti-HBc = hepatitis B core antibody; DLBCL = diffuse large B-cell lymphoma; HBV = hepatitis B virus; NHL = non-Hodgkin lymphoma; PR = partial response; ULN = upper limit of normal.

Source: L-MIND Clinical Study Report,¹⁰ L-MIND Clinical Study Protocol.²⁴

A summary of key protocol deviations during the L-MIND study is presented in Table 10. As of the primary analysis (November 30, 2018, data cut), diagnosis of DLBCL was not confirmed by the central pathology lab in 9 patients (11.1%). ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. The following subtypes of NHL were classified as non-DLBCL cases: follicular lymphoma (grade 2 + 3A), follicular lymphoma grade 2, mantle-cell lymphoma, classic type, and marginal zone lymphoma. Diagnosis remained unknown because there was insufficient tissue and because no biopsy material was available for a further 2 patients (2.5%). ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. Two patients with double- or triple-hit genetics were enrolled in L-MIND.

Table 10: Summary of Key Protocol Deviations – Safety Analysis Set, November 30, 2018, Data Cut-Off (Primary Analysis)

Source: L-MIND Clinical Study Report.¹⁰

Analysis Populations

The FAS included all patients who received at least 1 dose of tafasitamab and 1 dose of lenalidomide (i.e., both study drugs must have been received at least once).

The safety analysis set included all patients who received at least 1 dose of either study drug (tafasitamab or lenalidomide).

Results

Patient Disposition

Patient disposition in the L-MIND study (summarized in Table 11) was the same at each of the analyses (November 2018, November 2019, and October 2020 data cut-off dates). In total, 156 patients were screened, of whom 75 (48.1%) were screen failures. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| A total of 81 patients were enrolled in L-MIND, and 80 (98.8%) patients received both tafasitamab and lenalidomide (i.e., were included in the FAS). One enrolled patient received tafasitamab only and was excluded from the FAS. Overall, 30 (37.5%) patients successfully completed the combination treatment phase on both study drugs (12 cycles); 50 (62.5%) patients did not successfully complete the combination treatment phase. In total, 45 (56.3%) discontinued both study drugs before completing 12 cycles; the primary reasons were disease relapse (39.5%), AE (9.9%), withdrawal by patient (2.5%), death (2.5%), and other (2.5%). Four (4.9%) patients discontinued lenalidomide only due to AEs before completing combination treatment. One (1.2%) patient discontinued tafasitamab only due to an AE before completing combination treatment. Additional details on the reasons for study treatment discontinuation are presented in Table 12.

As of the October 30, 2020, data cut-off, 34 (42.0%) patients had reached cycle 13 day 1 (beginning of treatment with tafasitamab monotherapy). Fifteen (44.1%) of these patients treated with tafasitamab monotherapy had discontinued tafasitamab treatment as of the data cut-off date due to disease progression (n = 8 patients), withdrawal by patient (n = 4), other reasons (n = 2), or an AE (n = 1). Thus, 19 patients were continuing tafasitamab monotherapy, and 62 patients had discontinued study treatment at the data cut-off. Of the patients who had discontinued study treatment, 42 had died, 13 were in survival follow-up, and 7 were lost to follow-up at the data cut-off.

Table 11: L-MIND Patient Disposition

FAS = full analysis set.

^aOne enrolled patient received tafasitamab only and was excluded from the FAS.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Exposure to Study Treatments

Exposure to study treatment in the L-MIND study as of the primary analysis (November 30, 2018. data cut-off date) and most recent interim analysis (October 30, 2020, data cut-off) is summarized in Table 12. Exposure to study treatment was not reported at the interim analysis with the November 30, 2019, data cut-off date. All 81 (100%) patients enrolled in the L-MIND study received at least 1 dose of a study drug. As of the primary analysis and most recent analysis, the median duration of exposure to the study treatment (tafasitamab plus lenalidomide) was 9.2 months.

Concomitant medications were defined as medications with a start date after the start of the treatment period but before study completion or discontinuation. As of the primary analysis, all patients in the FAS reported the use of concomitant medications. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Dosage Interruptions and Reductions

As of the primary analysis (November 30, 2018, data cut-off), 61/81 (75.3%) patients had temporary interruptions of tafasitamab in total, including skipped visits and infusion interruptions. The most common reason for temporary interruption of tafasitamab was AEs (n = 47/61, 77.0%). As of the October 30, 2020, data cut-off, 64/81 (79.0%) patients had temporary interruptions of tafasitamab. The most common reason was AEs (n = 47/64, 73.4%). Dosage reductions of tafasitamab were not permitted per the study protocol.

As of both analyses, 28/80 (35.0%) patients who received lenalidomide had temporary interruptions. The most common reason was AEs, in 25 (89.3%) patients. Thirty-seven (45.7%) patients had at least 1 dosage reduction for lenalidomide.

Table 12: Summary of Exposure to Study Treatment in L-MIND – Safety Analysis Set (N = 81)

AE = adverse event; DCO = data cut-off; SD = standard deviation; Q1 = first quartile; Q3 = third quartile.

^aDuration of exposure to combination therapy or lenalidomide only was the time from first administration of study drug to the last date of administration of lenalidomide or the last date of exposure to tafasitamab during combination treatment.

Source: L-MIND Clinical Study Report¹⁰ and Addendum 3.¹⁴

Compliance

Compliance with tafasitamab was calculated by taking the actual tafasitamab infusion doses divided by the planned tafasitamab infusion doses. Compliance was summarized per single infusion and per cycle. A boxplot depicting compliance with tafasitamab by treatment cycle as of the primary analysis (November 30, 2018, data cut-off) is presented in Appendix 3 (Figure 9). As of the primary analysis, the median compliance for tafasitamab for each cycle was 100% until cycle 34.

Compliance with lenalidomide was calculated by taking the total use of lenalidomide doses divided by the total planned lenalidomide doses. If these values were 80% or more and 120% or less of the planned dosage for the cycle, then the patient was considered compliant for that cycle. A boxplot of compliance with lenalidomide by cycle as of the primary analysis is presented in Appendix 3 (Figure 10). As of the primary analysis, the median compliance for lenalidomide was 100% for each cycle.

Subsequent Anticancer Treatments

As of the primary analysis (November 30, 2018, data cut-off date), 25 (31.3%) patients received non-study anticancer treatment after discontinuing study treatment. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported in this section. See Appendix 3 for detailed efficacy data, including sensitivity analyses of patients with central pathology-confirmed DLBCL (Table 50).

Time-to-Event Outcomes

Overall Survival

A summary of OS results from the L-MIND as of the 3 data cut-off dates is presented in Table 13. As of the primary analysis (November 30, 2018, data cut-off), median follow-up time for OS was 19.6 months (95% CI, 15.3 to 21.9) and 29 (36.3%) deaths were observed in the FAS. The Kaplan–Meier plot of OS as of the primary analysis is depicted in Figure 3. The Kaplan–Meier estimate for the median OS was NR (95% CI, 18.3 to NR).

As of the October 30, 2020, data cut-off, median follow-up time for OS was 42.7 months (||||||||||||||||||||||||||||||||||||) and 41 (51.3%) deaths had occurred. Thirty-nine patients were censored in the OS analysis ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. The Kaplan–Meier plot of OS as of the October 30, 2020, data cut-off is depicted in Figure 4. The Kaplan–Meier estimate for the median OS was 33.5 months (95% CI, 18.3 to NR).

The results of subgroup analyses of OS are presented in Table 14.

Table 13: Summary of OS Results in L-MIND – FAS

CI = confidence interval; DCO = data cut-off; FAS = full analysis set; NR = not reached; OS = overall survival.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

^cCalculated using the reverse Kaplan–Meier method, considering the censored patients as events and patients with events as censored.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Figure 3: Kaplan–Meier Plot of OS in L-MIND – FAS; November 30, 2018, Data Cut-Off (Primary Analysis)

CI = confidence interval; FAS = full analysis set; LEN = lenalidomide; MOR00208 = tafasitamab; NR = not reached.

Source: L-MIND Clinical Study Report.¹⁰

Figure 4: Kaplan–Meier Plot of OS in L-MIND – FAS; October 30, 2020, Data Cut-Off

CI = confidence interval; FAS = full analysis set; LEN = lenalidomide; MOR00208 = tafasitamab; NR = not reached.

Source: L-MIND Clinical Study Report Addendum 3.¹⁴

Table 14: Subgroup Analyses of OS – FAS

ASCT = autologous stem cell transplantation; CI = confidence interval; DCO = data cut-off; FAS = full analysis set; GCB = germinal centre B-cell; HDT = high-dose chemotherapy; IHC = immunohistochemistry; IPI = International Prognostic Index; NR = not reached; OS = overall survival.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Progression-Free Survival

Results for PFS by IRC and PFS by investigator assessment in the L-MIND trial are summarized in Table 15.

As of the primary analysis (November 30, 2018, data cut-off), the median follow-up time for PFS by IRC in the FAS was 17.3 months (95% CI, 11.5 to 21.2) in the FAS. By IRC assessment, PFS events were observed in 39 (48.8%) patients. The Kaplan–Meier curve of PFS by IRC is depicted in Figure 5. The Kaplan–Meier estimate for the median PFS by IRC was 12.1 months (95% CI, 5.7 to NR). By investigator assessment, 44 (55.0%) patients had experienced an event and median PFS by investigator was 9.1 months (95% CI, 5.5 to 22.3).

As of the October 30, 2020, data cut-off date, the median follow-up time for PFS by IRC evaluation was 33.9 months (95% CI, 26.5 to 35.4) in the FAS. By IRC assessment, PFS events were observed in 42 (52.5%) patients. The Kaplan–Meier curve of PFS by IRC is depicted in Figure 6. The Kaplan–Meier estimate for median PFS by IRC was 11.6 months (95% CI, 6.3 to 45.7). By investigator assessment, 46 (57.5%) patients experienced a PFS event and median PFS by investigator was 9.1 (95% CI, 5.5 to 28.0) months.

Results of the subgroup analyses of PFS by IRC are reported in Table 16.

Table 15: Summary of PFS Results in L-MIND – FAS

CI = confidence interval; DCO = data cut-off; FAS = full analysis set; IRC = independent review committee; NR = not reached; PFS = progression-free survival.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

^cCalculated using the reverse Kaplan–Meier method, considering the censored patients as events and patients with events as censored.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Figure 5: Kaplan–Meier Plot of PFS by IRC in L-MIND – FAS; November 30, 2018, Data Cut-Off (Primary Analysis)

CI = confidence interval; FAS = full analysis set; IRC = independent review committee; LEN = lenalidomide; MOR00208 = tafasitamab; NR = not reached; PFS = progression-free survival.

Source: L-MIND Clinical Study Report.¹⁰

Figure 6: Redacted

This figure has been redacted.

Table 16: Subgroup Analyses of PFS by IRC – FAS

ASCT = autologous stem cell transplantation; CI = confidence interval; DCO = data cut-off; FAS = full analysis set; GCB = germinal centre B-cell; HDT = high-dose therapy; IHC = immunohistochemistry; IPI = International Prognostic Index; NR = not reached; PFS = progression-free survival.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Time to Progression

TTP based on IRC evaluation was analyzed at the primary analysis and results are presented in Table 17. As of the November 30, 2018, data cut-off, 35 (43.8%) patients had progressed or died and the median TTP was 16.2 months (95% CI, 7.4 to NR). Median follow-up time for TTP was not reported. TTP was not analyzed at the subsequent interim analyses.

Event-Free Survival

EFS was not a protocol-defined end point in L-MIND and thus is considered exploratory. EFS was analyzed at the primary analysis, and results are presented in Table 18. As of the November 30, 2018, data cut-off, 46 (57.5%) patients had experienced progression, death, or received a new non-study antineoplastic treatment. Median EFS was 9.1 (95% CI, 5.3 to 21.0) months. The median follow-up time for EFS was 19.7 months (95% CI, 14.3 to 22.0). EFS was not analyzed at the subsequent interim analyses.

Table 17: Summary of TTP by IRC Results in L-MIND – FAS; November 30, 2018, Data Cut-Off (Primary Analysis)

CI = confidence interval; FAS = full analysis set; IRC = independent review committee; NR = not reached; TTP = time to progression.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report.¹⁰

Table 18: Redacted

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Response/Remission Rate

Objective Response Rate

The ORR by IRC was the primary end point of the L-MIND study. The ORR by investigator assessment was a secondary end point. Results for ORR by IRC and ORR by investigator, including the proportion of patients who achieved CR (i.e., CRR) and PR, are presented in Table 19.

As of the primary analysis (November 30, 2018, data cut-off), the ORR by IRC was 60.0% (95% CI, 48.4 to 70.8). The best objective response for patients was CR for 34/80 (42.5%) patients and PR for 14/80 (17.5%) patients. Based on investigator assessment, the ORR was 63.8% (95% CI, 52.2 to 74.2).

As of the most recent efficacy analysis (October 30, 2020, data cut-off date), ORR by IRC was 57.5% (95% CI, 45.9 to 68.5). Thirty-two (40.0%) patients had CR and 14 (17.5%) patients had PR. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

At each analysis, 8 patients by IRC and 7 patients by investigator were considered not evaluable because they had no valid post-baseline assessment.

The results of the subgroup analyses of ORR by IRC are presented in Table 20.

Table 19: Summary of ORR Results in L-MIND – FAS

CI = confidence interval; CR = complete response; DCO = data cut-off; FAS = full analysis set; IRC = independent review committee; ORR = objective response rate; PD = progressive disease; PR = partial response; SD = stable disease.

^a95% CI calculated using 2-sided 95% Clopper-Pearson exact method based on binomial distribution.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Table 20: Subgroup Analyses of ORR by IRC (Primary End Point) – FAS

ASCT = autologous stem cell transplantation; CI = confidence interval; DCO = data cut-off; FAS = full analysis set; GCB = germinal centre B-cell; HDT = high-dose therapy; IHC = immunohistochemistry; IPI = International Prognostic Index; IRC = independent review committee; n = patients with response; ORR = objective response rate.

^a95% CI calculated using 2-sided 95% Clopper-Pearson exact method based on binomial distribution.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Duration of Response

A summary of DOR by IRC and DOR by investigator assessment is presented in Table 21. Kaplan–Meier plots of DOR by IRC assessment by best objective response achieved (CR or PR) as of the primary analysis and most recent analysis are depicted in Figure 7 and Figure 8, respectively.

As of the primary analysis (November 30, 2018, data cut-off), 48 patients had a response (CR or PR) by IRC assessment. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The Kaplan–Meier estimate for the median DOR by IRC was 21.7 (95% CI, 21.7 to NR) months. Median DOR by IRC in patients with PR was 4.4 months (95% CI, 2.0 to 9.1) and NR (95% CI, 21.7 to NR) in patients with CR.

As of the most recent analysis (October 30, 2020, data cut-off), 46 patients had a response by IRC assessment. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The Kaplan–Meier estimate for the median DOR by IRC was 43.9 (95% CI, 26.1 to NR) months. Median DOR by IRC in patients with PR was 5.6 (95% CI, 2.2 to NR) months compared to NR (95% CI, 43.9 to NR) in patients with CR.

Results of the subgroup analyses for DOR by IRC are presented in Table 22.

Table 21: Summary of DOR Results in L-MIND – FAS

CI = confidence interval; DCO = data cut-off; DOR = duration of response; FAS = full analysis set; IRC = independent review committee; NR = not reached.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Figure 7: Kaplan–Meier Plot of DOR by IRC by Best Objective Response in L-MIND – FAS; November 30, 2018, Data Cut-Off (Primary Analysis)

CR = complete response; DOR = duration of response; FAS = full analysis set; IRC = independent review committee; LEN = lenalidomide; NR = not reached; PR = partial response.

Source: L-MIND Clinical Study Report.¹⁰

Table 22: Subgroup Analyses of DOR by IRC – FAS

ASCT = autologous stem cell transplantation; CI = confidence interval; DCO = data cut-off; DOR = duration of response; FAS = full analysis set; GCB = germinal centre B-cell; HDT = high-dose therapy; IHC = immunohistochemistry; IPI = International Prognostic Index; IRC = independent review committee; NR = not reached.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Time to Response

Results for TTR as of the primary analysis and November 30, 2019, data cut-off analysis are summarized in Table 23. TTR data were not reported at the most recent interim analysis (October 30, 2020, data cut-off).

As of the primary analysis (November 30, 2018, data cut-off), median TTR (CR or PR) based on IRC evaluation was 2.0 months (range 1.7 to 16.8 months). Median time to PR was 1.90 months (range 1.7 to 16.8 months). Median time to CR was 7.05 months (range 1.7 to 17.0 months). Median time to best response (either CR or PR) was 3.6 months (range 1.7 to 17.0 months).

As of the November 30, 2019, data cut-off, median TTR based on IRC evaluation was 2.0 months (range 1.7 to 16.8). Median time to CR was 4.0 months (range 1.7 to 17.0). Median time to PR and median time to best response were not reported.

Figure 8: Kaplan–Meier Plot of DOR by IRC by Best Objective Response in L-MIND – FAS; October 30, 2020, Data Cut-Off

CR = complete response; DOR = duration of response; FAS = full analysis set; IRC = independent review committee; LEN = lenalidomide; NR = not reached; PR = partial response.

Source: L-MIND Clinical Study Report Addendum 3.¹⁴

Time to Next Treatment

Results for TTNT as of the primary analysis and November 30, 2019, data cut-off analysis are summarized in Table 24. TTNT was not reported at the most recent interim analysis (October 30, 2020, data cut-off). Median TTNT was 15.4 (95% CI, 7.6 to NR) months at the primary analysis compared to 12.5 (95% CI, 7.6 to 24.7) at the November 2019 interim analysis. Median follow-up time for TTNT was not reported.

Results from a subgroup analysis of TTNT by IPI score conducted at the primary analysis are presented in Table 25. No other subgroup analyses for TTNT were reported.

Table 23: Summary of TTR by IRC Results in L-MIND – FAS

CR = complete response; DCO = data cut-off; FAS = full analysis set; IRC = independent review committee; PR = partial response; TTR = time to response.

Source: L-MIND Clinical Study Report,¹⁰ Addendum 1,¹² and Addendum 3.¹⁴

Table 24: Summary of TTNT Results in L-MIND – FAS, November 30, 2018, Data Cut-Off (Primary Analysis)

CI = confidence interval; DCO = data cut-off; FAS = full analysis set; NR = not reached; TTNT = time to next treatment.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report¹⁰ and Addendum 1.¹²

Table 25: Subgroup Analysis of TTNT by IPI Score – FAS, November 30, 2018, Data Cut-Off (Primary Analysis)

CI = confidence interval; FAS = full analysis set; IPI = International Prognostic Index; NR = not reached; TTNT = time to next treatment.

^aKaplan–Meier estimate.

^b95% CI calculated using the Brookmeyer and Crowley (1982) method.

Source: L-MIND Clinical Study Report.¹⁰

Health-Related Quality of Life

HRQoL outcomes were not reported in the L-MIND study.

Harms

Only those harms identified in the review protocol are reported below. See Table 26 for detailed harms data for the safety analysis set as of the most recent analysis of the L-MIND trial (data cut-off date October 30, 2020). Safety data after the longer follow-up was similar to safety data presented at the primary analysis (November 30, 2018, data cut-off). All 81 (100%) enrolled patients received at least 1 dose of the study drug and were included in the safety analysis set.

Adverse Events

As of the October 30, 2020, data cut-off, all 81 (100%) patients enrolled in L-MIND experienced 1 or more treatment-emergent AE. The most common AEs were neutropenia (50.6%), anemia (37.0%), diarrhea (35.8%), thrombocytopenia (30.9%), and cough (27.2%).

Serious Adverse Events

Overall, 53.1% of patients enrolled in L-MIND experienced 1 or more SAEs as of the October 30, 2020, data cut-off. The most common SAEs were pneumonia (n = 7, 8.6%), febrile neutropenia (n = 5, 6.2%), and pulmonary embolism (n = 3, 3.7%). Other SAEs reported in more than 1 patient included bronchitis, lower respiratory tract infection, atrial fibrillation, and congestive cardiac failure (n = 2, 2.5% each).

Withdrawal Due to Adverse Events

Overall, 20 (24.7%) patients permanently discontinued treatment with 1 or both study drugs due to AEs: 8 (9.9%) patients discontinued lenalidomide only, 2 (2.5%) discontinued tafasitamab only, and 10 (12.3%) discontinued both study drugs. The only AE that led to permanent discontinuation of study drug in more than 1 patient was neutropenia (n = 3, 3.7%).

Mortality

In total, 42 (51.9%) patients enrolled in L-MIND had died as of the October 30, 2020, data cut-off date. The cause of death was reported to be related to disease progression for 31 (38.3%) patients and unrelated to disease progression in 10 (12.3%) patients.

Notable Harms

Overall, 72.8% of patients enrolled in L-MIND experienced an infection. The most common types of infections were bronchitis (16.0%), pneumonia (12.3%), urinary tract infection (12.3%), and respiratory tract infection (11.1%).

Regarding myelosuppression, 50.6% of patients experienced neutropenia, 37.0% experienced anemia, 30.9% experienced thrombocytopenia, 14.8% experienced leukopenia, 12.3% experienced febrile neutropenia, and 7.4% experienced lymphopenia.

One (1.2%) patient developed worsening PML. |||||||||||||||||||||||| experienced hepatitis B reactivation. Five (6.2%) patients experienced an infusion-related reaction. No patients experienced a grade 3 or higher tumour lysis syndrome or cytokine release syndrome. Tumour lysis syndrome or cytokine release syndrome events of any grade were not reported.

Table 26: Summary of Harms – Safety Analysis Set; October 30, 2020, Data Cut-Off Date

AE = adverse event; PML = progressive multifocal leukoencephalopathy; SAE = serious adverse event.

^aFrequency more than 15%.

^bOccurred in more than 1 patient.

^cFrequency more than 5%.

^dNo patients experienced a grade 3 or higher tumour lysis syndrome or cytokine release syndrome. Tumour lysis syndrome or cytokine release syndrome events of any grade were not reported.

Source: L-MIND Clinical Study Report Addendum 3.¹⁴

Critical Appraisal

Internal Validity

For the primary end point and multiple secondary end points (i.e., PFS, EFS, DOR, TTR), an IRC was appropriately used to reduce the risk of detection bias. The IRC consisted of independent radiology and hemato-oncology physicians. Furthermore, there was generally good agreement between the IRC and investigator-assessed outcomes. The criteria used to assess response (IWG response criteria reported by Cheson et al. [2007])⁸ were appropriate, although the clinical experts consulted by CADTH noted that there are more recent guidelines available (i.e., Lugano criteria)³³ that include PET scanning criteria more extensively. It is possible that the results could have differed if other criteria had been used, although the extent is uncertain.

The study population in L-MIND was adequately defined, and the clinical experts consulted by CADTH indicated that the eligibility criteria were appropriate. Approximately half of the patients screened were screen failures. Screen failures were driven predominantly by laboratory parameters exclusion criteria, which were considered necessary by the sponsor due to the starting dosage of lenalidomide. Study treatment discontinuation rates were as expected by the clinical experts. It was noted that most discontinuations of study drug(s) were due to disease relapse, which was expected by the clinical experts.

The analysis populations used in the L-MIND trial were appropriate. The efficacy outcomes were analyzed descriptively in the FAS population, which excluded 1 enrolled patient. Exclusion of 1 patient is unlikely to affect outcomes. Safety outcomes were assessed in all patients who were treated with a study drug.

The primary analysis (November 30, 2018, data cut-off date) was pre-specified in the L-MIND study protocol. The sample size of 80 patients was adequate to estimate the primary end point—ORR by IRC—with high statistical precision. The L-MIND trial met its primary end point (i.e., improvement of ORR as per IRC from 20% to 35% with single-drug therapy) at the primary analysis, as there was an ORR by IRC of 60.0% (95% CI, 48.4 to 70.8). The interim analyses with November 30, 2019, and October 30, 2020, data cut-off dates were not pre-specified. The November 30, 2019, data cut-off analysis was used in the Health Canada; the October 30, 2019, data cut-off analysis was conducted to report long-term outcomes after greater follow-up time. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

L-MIND is an open-label, single-arm study. There is no direct evidence comparing tafasitamab plus lenalidomide to a control arm. Furthermore, no statistical testing was performed because the L-MIND study was not designed to test hypotheses. Data were analyzed descriptively. Due to these limitations of the study design, no definitive conclusions can be drawn by the CADTH review team from the L-MIND study regarding the efficacy and safety of tafasitamab plus lenalidomide relative to a relevant comparator. The results were not adjusted for potential confounders (i.e., effect modifiers or prognostic factors) that could affect outcomes. The open-label design can increase the risk of performance and detection bias, particularly for outcomes that are subjective in measurement and interpretation (e.g., response, AEs). There is a high risk of performance and detection bias for known subjective harms, which could be overestimated, since both patients and their treating clinicians knew the treatment received and knew that they were participating in a trial. Objective outcomes such as OS time and mortality are unlikely to be affected by performance or detection bias. The potential for detection bias was minimized by using IRC assessment for key study outcomes such as ORR, DOR, and PFS.

The time-to-event analyses were appropriate, but causality cannot be inferred from a single-arm trial without a comparator. Survival times (median OS and median PFS) were estimated from the Kaplan–Meier models, and patients who did not have an event were censored, which is appropriate. The L-MIND trial is ongoing. Although the planned primary analysis has been conducted (data cut-off date of November 30, 2018), the final analysis has not yet been conducted. As of the October 30, 2020, data cut-off date, 19 (23.5%) patients were ongoing with tafasitamab monotherapy treatment. Per the Clinical Study Report, PFS data were considered mature as of the primary analysis, although numerical changes in median PFS at the subsequent analyses were noted. The median duration of follow-up at the time of the analyses was sufficient for other time-to-event end points (e.g., OS, DOR).

Most of the subgroup analyses of OS, PFS, and DOR were pre-specified, except for the subgroup analyses of IPI score. Sample sizes for the subgroup analyses were small, and the CIs reflected imprecision. In addition, the subgroup analyses were exploratory, and there were no statistical comparisons. As a result, the CADTH review team can draw no conclusions based on the subgroup analyses.

Multiple protocol amendments were implemented while the L-MIND study was being conducted, which included changes to the study eligibility criteria. In the original trial protocol, patients were required to have a histologically confirmed diagnosis of DLBCL not otherwise specified. Patients with NHL other than classical histology DLBCL (e.g., including patients with DLBCL transformed from indolent lymphomas) were excluded. In addition, patients who had relapsed within 3 months of prior CD20-targeted therapy were excluded. The first protocol amendment, which was implemented before patient enrolment began, expanded the study eligibility criteria to include patients with evidence of histological transformation to DLBCL from indolent NHL. Since this change was implemented before study patients were enrolled, it is less likely to have introduced bias. In subsequent protocol amendments, which were implemented after participant enrolment commenced, the eligibility criteria were changed to allow up to 3 lines of prior therapy for DLBCL treatment (previously, 2 prior lines were allowed), to remove the upper age limit of 80 years for study entry, and to allow patients with Gilbert’s syndrome or liver involvement by lymphoma. Changes in study eligibility after enrolment had begun may introduce selection bias, although the direction of bias is unknown. The clinical experts consulted by CADTH noted that the changes in eligibility criteria generally made the trial population more closely resemble the patient population in Canada with R/R DLBCL who are not eligible for ASCT.

In the original trial protocol, HBV serology was required monthly for all patients. An amendment implemented after participant enrolment had begun changed the requirement to HBV serology monthly only for those patients who were anti-HBc antibody–positive and HBV-DNA–negative at screening. This may have biased the detection of HBV reactivation, leading to underestimation of this notable harm.

In the original trial protocol, treatment with tafasitamab was continued until a maximum of 24 cycles (i.e., approximately 2 years). In the last protocol amendment, treatment with tafasitamab was extended beyond cycle 24 until disease progression. The reason for extending the treatment duration was unclear.

A high rate of protocol deviations occurred in L-MIND, which creates uncertainty in the data. Key protocol deviations were related to procedures or tests, study drug and treatment, laboratory assessment, informed consent form, eligibility criteria, and prohibited concomitant mediations. Deviations in study drug and treatment could have biased the efficacy and safety results, thus affecting internal validity. Deviations in eligibility criteria may also have resulted in selection bias. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Of note, the L-MIND study consisted of patients with R/R DLBCL diagnosed as per local pathology. However, central pathologic analysis concluded that approximately 10% of these patients had non-DLBCL histology or alternative diagnoses. The following subtypes of NHL were classified as non-DLBCL cases: follicular lymphoma (grade 2 + 3A), follicular lymphoma grade 2, mantle-cell lymphoma, classic type, and marginal zone lymphoma. The clinical experts consulted by CADTH indicated that patients could be misdiagnosed due to numerous factors (e.g., the biopsy, expertise in lymphoma). However, enrolment of these patients introduced bias because these patients are not representative of the intended patient population. Patients with alternative diagnoses may have had a better prognosis than patients with R/R DLBCL, thus overestimating the treatment effects of tafasitamab plus lenalidomide. Sensitivity analyses were performed on patients who had DLBCL confirmed by central pathologic analysis versus those who had not. Efficacy results for the patients with central pathology-confirmed DLBCL were generally consistent with the FAS.

In L-MIND, flattening of the OS and PFS Kaplan–Meier curves was observed. The clinical experts reported that they do not see flattening of OS and PFS curves with other available treatments in patients with R/R DLBCL who are not eligible for ASCT. The clinical experts indicated that subsequent anticancer treatments may have confounded the PFS and OS results. Furthermore, i a small number of patients were at risk at the tail end of the Kaplan–Meier curves, and these results may have been confounded by the patients with histologies other than DLBCL.

Overall, the L-MIND study was a phase II trial that enrolled 80 patients in the FAS. The clinical experts consulted by CADTH indicated that it may not be possible to extrapolate efficacy results from this small sample of patients to the general population of patients with R/R DLBCL in Canada. The clinical experts indicated that results should be confirmed in a phase III trial with a comparator arm.

External Validity

The L-MIND trial was an international, multi-centre study that included sites in Europe and the US. There were no study sites in Canada. The treatment regimen used in the L-MIND trial aligns with the Health Canada–recommended dose of tafasitamab plus lenalidomide. In the L-MIND study, pre-medications as prophylaxis for infusion-related reactions were administered before tafasitamab infusions, which was appropriate according to the clinical experts. The clinical experts indicated that pre-medications are commonly used in standard practice to prevent infusion-related reactions with other monoclonal antibody treatments.

The clinical experts consulted by CADTH reported that standard of care for assessing treatment response would be imaging with CT-PET every 3 to 4 months with clinical examination and bloodwork before each treatment. In the L-MIND study, imaging to assess disease response was conducted every 2 cycles (i.e., approximately every 2 months) during the 12 months of combination treatment. While the patients received tafasitamab monotherapy, imaging was conducted every 3 months from cycle 13 to 24, then approximately once every year from cycle 25 onward.

The clinical experts indicated that the baseline characteristics of patients enrolled in L-MIND were generally representative of the R/R DLBCL patient population in Canada, although they noted that the L-MIND study patients represent the most fit patients in this population, which is common in clinical trials. Some differences between the L-MIND study population and population of patients in Canada

with R/R DLBCL who are ineligible for ASCT were noted. The clinical experts indicated that the study population was younger than the general population of patients with R/R DLBCL who are eligible for ASCT, as many patients are ineligible due to older age (i.e., > 70 or > 75 years, depending on the site). According to the clinical experts, the population of patients in Canada with R/R DLBCL has a greater proportion of patients with an ECOG PS of 2, more patients have a relapse within 12 months of initial therapy, more patients have primary refractory disease, and more patients have had a prior ASCT that failed compared to the L-MIND study population. Furthermore, the large rate of screen failure indicates that many patients with R/R DLBCL were not eligible for participation in L-MIND, thus limiting generalizability if a large proportion of patients normally seen in clinical practice in Canada would not have been eligible.

Overall, the clinical experts thought that the eligibility criteria used in the L-MIND study were appropriate and allowed enrolment of patients with R/R DLBCL who were generally representative of the patient population in Canada. However, it noted that some groups of patients in the R/R DLBCL patient population were excluded. The L-MIND study excluded patients with known double- or triple-hit genetics DLBCL at study entry, which limits the generalizability of results to this patient population. In addition, primary refractory DLBCL was an exclusion criterion in the L-MIND study. The initial definition led to exclusion of relapses within 3 months of a prior anti-CD20 therapy. In a later amendment of the protocol, the definition of primary refractory DLBCL was revised to less than a PR to first-line therapy or progression within 6 months after completion of first-line therapy, and removed the need to have DLBCL relapse or progression after at least 3 months from completion of prior anti-CD20–containing therapy. As a result of this change in definition, 15 (18%) of patients enrolled in the L-MIND trial were considered to have primary refractory disease. This complicates the interpretation of the generalizability of study results to the population of patients who have primary refractory disease. Because of the change in definition and because primary refractory disease is an exclusion criterion in the pivotal study, the clinical experts consulted by CADTH indicated that it was unclear whether results were generalizable to patients with primary refractory disease.

Multiple protocol amendments related to the eligibility criteria were implemented during the L-MIND study and affect the generalizability of results. The first protocol amendment, implemented before patient enrolment began, expanded the study eligibility criteria to include patients with evidence of histological transformation to DLBCL from indolent NHL. In subsequent protocol amendments, implemented after participant enrolment commenced, the eligibility criteria were changed to allow up to 3 lines of prior therapy for DLBCL treatment (previously, 2 prior lines were allowed), to remove the upper age limit of 80 years for study entry, and to allow patients with Gilbert’s syndrome or liver involvement by lymphoma. According to the clinical experts, these changes to the study eligibility criteria increased the generalizability of results, as they made the trial population more representative of the general patient population in Canada.The clinical experts indicated that the criteria used to defined ASCT ineligibility in the L-MIND study were reasonable. The L-MIND study defined older age as > 70 years old as part of the ineligibility for ASCT criteria. The clinical experts thought that this was reasonable, although they noted that some centres in Canada perform ASCTs in patients up to 75 years of age. The clinical experts noted that organ function and predicted toxicities related to the transplant are important in determining a patient’s eligibility for ASCT.

The outcomes OS, PFS, ORR, and DOR are commonly used in clinical trials of anticancer therapy and are relevant to clinical practice. The patient groups that provided input on this review indicated that longer survival, remission, and controlling disease symptoms were most important outcomes for a new therapy. Better HRQoL and fewer side effects compared to current therapies were also important considerations for patients. Similarly, the clinical experts indicated that a clinically meaningful response to treatment would include improvement in survival and DOR, which they would expect to correlate with improvement in symptom burden. According to the clinical experts, meaningful response would include CR, PR, or stable disease with a tolerable toxicity profile. The L-MIND trial assessed some outcomes that were important to patients (e.g., OS, PFS, EFS, and treatment-emergent AEs). Disease symptoms and HRQoL were not reported in L-MIND, which represents a gap in the evidence.

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

Because there was no direct evidence comparing tafasitamab plus lenalidomide to other therapies for the treatment of R/R DLBCL, a review of the indirect evidence was undertaken. In addition to reviewing the sponsor’s submission, CADTH conducted a literature search to identify potentially relevant ITCs in patients with DLBCL. A focused literature search for ITCs dealing with DLBCL was run in MEDLINE All (1946–) on December 15, 2021. No date or language limits were applied. Titles, abstracts, and full-text articles were screened for inclusion by 1 reviewer based on the population, intervention, comparator, and outcome criteria outlined in Table 5.

No relevant ITCs were identified in the literature search. Three sponsor-submitted indirect comparisons were included in this review: 2 studies in which retrospective observational cohorts were compared to patients enrolled in L-MIND using ePS-based NN 1:1 matching methodology and unanchored MAICs. These ITCs were used to inform the pharmacoeconomic models.

Description of Indirect Comparison(s)

The RE-MIND^16,17 and RE-MIND2¹⁸ studies were retrospective observational studies conducted to generate an external control for indirect comparison with the L-MIND cohort. A summary of the key study design features and analysis methods of RE-MIND and RE-MIND2 is provided in Table 27.

Table 27: RE-MIND and RE-MIND2 Study Characteristics and Analysis Methods

AE = adverse event; ANC = absolute neutrophil count; ASCT = autologous stem cell transplant; BR = bendamustine plus rituximab; CAR T-cell = chimeric antigen receptor T-cell; CNS = central nervous system; CRR = complete response rate; CT = CT; DCR = disease control rate; DLBCL = diffuse large B-cell lymphoma; DOR = duration of response; EBV = Epstein-Barr virus; EFS = event-free survival; ePS = estimated propensity score; ESMO = European Society for Medical Oncology; FAS = full analysis set; Hb = hemoglobin; HDC = high-dose chemotherapy; ITC = indirect treatment comparison; IWG = International Working Group; LDH = lactate dehydrogenase; MAS 25 = matched analysis set with starting dosage 25 mg per day; MAS_BR = matched analysis set for tafasitamab plus lenalidomide and bendamustine plus rituximab; MAS_Pool = matched analysis set for tafasitamab plus lenalidomide and pooled systemic therapies; MAS_R-GemOx = matched analysis set for tafasitamab plus lenalidomide and rituximab plus gemcitabine plus oxaliplatin; MRI = MRI; NCCN = National Comprehensive Care Network; NN = nearest neighbour; NOS = not otherwise specified; ORR = objective response rate; PET = PET; PFS = progression-free survival; pola-BR = polatuzumab vedotin plus bendamustine plus rituximab; R-GemOx = rituximab plus gemcitabine plus oxaliplatin; R/R = relapsed or refractory; THRLBCL = T-cell or histiocyte-rich large B-cell lymphoma; TTNT = time to next treatment; ULN = upper limit of normal.

Source: RE-MIND Clinical Study Report,¹⁶ RE-MIND2 Clinical Study Report.¹⁸

RE-MIND

RE-MIND^16,17 was an international, retrospective, observational cohort study designed to characterize the effectiveness of lenalidomide monotherapy in the treatment of R/R DLBCL patients who were not eligible for HDC followed by ASCT. The effectiveness of tafasitamab plus lenalidomide in L-MIND was compared to lenalidomide monotherapy in an observational retrospective cohort study. In total, data were collected from 490 patients treated with lenalidomide and 140 qualified for matching with the L-MIND cohort. Overall, 76 eligible RE-MIND patients were identified and matched 1:1 to 76 L-MIND patients based on baseline characteristics. Data from the L-MIND study were from the November 30, 2018, data cut-off (i.e., the primary analysis).

RE-MIND2

RE-MIND2¹⁸ was an international, retrospective, observational cohort study designed to characterize the effectiveness of systemically administered therapies in the treatment of R/R DLBCL patients (second, third, and fourth line). Eligible systemic therapies included regimens administered in routine clinical care according to NCCN or ESMO guidelines^19,20 for patients who were not eligible for ASCT. The effectiveness of systemically administered therapies was then compared using the data collected retrospectively with the efficacy outcomes of tafasitamab plus lenalidomide in the L-MIND study after the cohorts were balanced. This study included the following treatment cohorts: systemic therapies pooled, BR, R-GemOx, rituximab plus lenalidomide, CAR T-cell therapy, pola-BR, or pixantrone monotherapy. Rituximab plus lenalidomide and pixantrone monotherapy are not relevant comparators in Canada, according to the clinical experts consulted by CADTH. Therefore, results for these cohorts will not be presented in the CADTH Clinical Review Report. Data used from L-MIND were from the second analysis (i.e., November 30, 2019, data cut-off). In total, 3,454 patients were enrolled in the observational study. In the pre-specified analysis, systemic therapies pooled, BR, and R-GemOx were compared to tafasitamab plus lenalidomide. Exploratory, post hoc analyses were conducted for the comparisons to pola-BR and CAR T-cell therapy. A propensity score–based, 1:1 matched comparison was conducted between the real-world data collected in the retrospective observational study and clinical trial data from L-MIND.

MAICs

Unanchored MAICs of tafasitamab plus lenalidomide in the L-MIND study versus comparators using prospective studies were conducted. The authors conducted a systematic literature review to identify relevant sources of evidence for comparator treatments for patients with R/R DLBCL who were not eligible for ASCT. In total, |||||||||||| prospective studies reporting data for ||||||||||||||||||||||||, pola-BR, BR, and R-GemOx were selected for the MAICs against tafasitamab plus lenalidomide. Data were used from the most recent interim analysis of the L-MIND study (i.e., October 30, 2020, data cut-off).

Methods of RE-MIND and RE-MIND2

Objectives

The primary objective of the RE-MIND study was to characterize the effectiveness of lenalidomide monotherapy compared with tafasitamab plus lenalidomide combination therapy in the treatment of R/R DLBCL patients.

The primary objective of the RE-MIND2 study was to compare the efficacy outcomes of the L-MIND cohort with the effectiveness in a matched patient population treated with systemic regimens administered in routine clinical care and listed in the NCCN or ESMO guidelines.^19,20

Populations

RE-MIND

Eligibility criteria for the RE-MIND observational cohort were aligned with the L-MIND study. Eligible patients were aged 18 years or older with histologically confirmed DLBCL, R/R after 1 to 3 prior systemic therapies (including 1 or more CD20-targeting regimens), and were not candidates for HDC and subsequent ASCT. Exclusion criteria included CNS lymphoma involvement, receiving lenalidomide in combination with another anti-lymphoma therapy (including radiation), prior treatment with anti-CD19 therapy or immunomodulatory imide drugs (e.g., thalidomide or lenalidomide), known double-hit or triple-hit genetics DLBCL, or a prior history of malignancies other than DLBCL (unless disease free for ≥ 5 years). Study sites were selected in Europe and the US (i.e., according to geographic distribution in L-MIND) based on data completeness and number of available patients. Eligible patients were identified from patient health records. The RE-MIND observational cohort did not have eligibility criteria related to ECOG PS or laboratory values.

RE-MIND2

Eligibility criteria for the RE-MIND2 observational cohort were also based on the L-MIND patient population. Eligible patients were aged 18 years or older at the initial DLBCL diagnosis; had a histologically confirmed diagnosis of DLBCL not otherwise specified, T-cell or histiocyte-rich large B-cell lymphoma, Epstein-Barr virus–positive DLBCL of the elderly, grade 3b follicular lymphoma, or composite lymphoma with a DLBCL component and a subsequent DLBCL relapse. Additionally, patients with the evidence of histological transformation to DLBCL from an earlier diagnosis of low-grade lymphoma into DLBCL with a subsequent DLBCL relapse were eligible. Patients were required to have R/R DLBCL and to have received at least 2 systemic regimens for the treatment of DLBCL, including at least 1 anti-CD20–containing therapy. Patients were excluded from the RE-MIND2 observational cohort if they had CNS involvement by lymphoma at initial DLBCL diagnosis, were treated with CD19-targeted therapy or immunomodulatory drugs (e.g., thalidomide, lenalidomide) as frontline DLBCL therapy, had undergone an allogenic stem cell transplant, had a prior history of malignancies other than DLBCL (unless the patient had been free of the disease for ≥ 5 years before inclusion), or had received tafasitamab. The RE-MIND2 observational cohort did not have eligibility criteria related to ECOG PS or laboratory values.

Eligible patients were identified from patient health records from sites selected across Europe, North America, and the Asia Pacific region, based on data completeness and number of available patients. Potential sites were identified from sponsor and contract research organization databases, as well as using input from external medical experts in the DLBCL indication. A Site Identification Questionnaire was designed and rolled out to the potential sites to obtain information on availability of eligible R/R DLBCL patients as well as on the availability and completeness of key variables. A site was selected for study participation if the availability of eligible R/R DLBCL patients and presence of key information were confirmed.

Recommended therapies for R/R DLBCL per NCCN or ESMO guidelines^19,20 eligible for inclusion in the RE-MIND2 observational cohort are listed in Appendix 3 (Table 52).

Outcomes

In both RE-MIND and RE-MIND2, the IWG response criteria reported by Cheson et al. (2014),³³ Cheson et al. (2007),⁸ Cheson et al. (1999),³⁴ or other criteria could be used for physician assessments of tumour response for the observational cohort. The following imaging modalities could be used to assess tumours: no radiological assessment done, PET-CT, PET-MRI, PET only, MRI only, CT only, and other. The interval between 2 tumour disease assessments was expected to be approximately 3 months (plus or minus 14 days).

RE-MIND

The primary end point for RE-MIND was ORR. Other end points included CRR, OS, DOR, PFS, TTNT, and EFS.

Best ORR was defined as the proportion of patients for whom CR or PR was the best response achieved at any time within ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

CRR was defined as the proportion of patients for whom CR was the best response achieved at any time within ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

OS was defined as the time from the index date until death from any cause (documented by the time of death). ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

DOR was defined as the elapsed time between the date of first documented response (CR or PR) and the date of an event defined as the first documented progression or death (from any cause).

PFS was defined as the time from the index date to the date of tumour progression or death from any cause. The date of progression was the first date for which PD was assessed as the objective response. PD was used as documented by the treating physician. Disease progression qualifying for a PFS event had to be confirmed by radiology assessment, bone marrow aspiration with confirmed lymphoma involvement, or tissue biopsy with confirmed lymphoma infiltration. Patients who did not experience an event were censored.

TTNT was defined as the time from index date to the start of next anti-DLBCL therapy ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

EFS was defined as the time from the index date to the date of disease progression, initiation of a new non-study anti-DLBCL treatment, or death ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

RE-MIND2

The primary end point for RE-MIND 2 was OS. Secondary end points included ORR, CRR, DOR, EFS, PFS, TTNT, treatment discontinuation rate due to AEs, and duration of treatment exposure.

OS was defined as the time from the index date for a given line until death from any cause (documented by the date of death). ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

ORR was defined as the proportion of patients for whom CR or PR was the best response achieved at any time within the analysis window or between index date for a given line and date of initiation of a new anti-DLBCL medication.

CRR was defined as the proportion of patients for whom CR was the best response achieved at any time within the analysis window or between index date for a given line and date of initiation of a new anti-DLBCL medication or death.

DOR was defined as the elapsed time between the date of first documented response (CR or PR) for a given line and the date of event defined as the first documented progression or death (from any cause).

PFS was defined as the time from the index date for a given line to the date of tumour progression or death from any cause. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

TTNT was defined as the time from index date for a given line to the start of next anti-DLBCL therapy (therapies included ASCT, a new systemic anti-DLBCL medication, surgery, or radiotherapy for any reason) or death due to any cause, whichever came first. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

EFS was defined as the time from the index date to the date of disease progression, initiation of a new anti-DLBCL treatment, or death from any cause, whichever came first.

ITC Analysis Methods

RE-MIND

Data were collected retrospectively from health records of patients in either real-world, compassionate use, and/or clinical trial settings. The effectiveness of lenalidomide monotherapy was then compared using the data collected retrospectively with the efficacy outcomes of tafasitamab plus lenalidomide combination therapy in the L-MIND study after the non-randomized cohorts were balanced on relevant, observable variables. Safety data were not collected except to document the reason for change in lenalidomide treatment.

Multiple sensitivity analyses were conducted. A sensitivity analysis was performed by applying NN 1:1 matching with caliper to ensure a high degree of balance in a subgroup of patients. The biggest caliper ensuring a standardized mean difference (SMD) of less than 0.2 for all 9 covariates involved in 1:1 matching was chosen. For residual imbalance after NN 1:1 matching for 1 or more covariates, a sensitivity analysis with doubly robust estimation was performed by adding the covariates with SMD of more than 0.2 following NN 1:1 matching in the respective logistic (for ORR and DCR) or Cox proportional hazard models (for PFS, OS, EFS, and TTNT).

A 6-month follow-up rule was implemented to prevent an overestimation of the rate of nonresponders in the RE-MIND observational cohort. A minimum of 6 months’ follow-up time was met if a patient responded (CR or PR) or progressed or died within 6 months from index date (from study day 1 to 183), or a responding patient (CR or PR as best response during analysis window) had a baseline tumour assessment and at least 1 post-baseline response assessment available at 6 months or later (on or after study day 184), or any patient who had at least 1 disease response assessment with SD, “indeterminate,” “not evaluable,” or “other” within 6 months from index date (from study day 1 to 183), with at least 1 assessment or death at 6 months or later (on or after study day 184).

The maximum follow-up time for an individual patient in the L-MIND study in the data cut-off used for the RE-MIND analyses was 32 months (first patient enrolled in March 2016, primary completion cut-off date November 30, 2018). To ensure a comparable distribution of follow-up times of patients in the observational study, an analysis window was applied, defined from index date to 32 months (974 days). This analysis window was applied for all analyses.

Statistical Analysis

Estimated propensity score–based NN 1:1 matching methodology was used to balance cohorts using the following 9 baseline covariates to estimate the propensity score: age (as categorical variable with subgroups < 70 versus ≥ 70 years of age); Ann Arbor stage I-II versus III-IV; refractoriness to last therapy line (yes versus no); number of prior lines of therapy (1 versus 2 or 3); history of primary refractoriness (yes versus no); prior ASCT (yes versus no); elevated lactate dehydrogenase (LDH; LDH > ULN versus LDH ≤ ULN); neutropenia (absolute neutrophil count < 1.5 × 10⁹/L versus absolute neutrophil count ≥ 1.5 × 109/L); anemia (hemoglobin < 10 g/dL versus ≥ 10 g/dL). A sensitivity analysis using imputation of missing data in baseline covariates was also performed.

For all end point models, a doubly robust estimation was conducted, in which 1 or more covariates were added to the end point estimation model to address any residual imbalance. Covariates added for doubly robust estimation were those with SMD of less than 0.20 following balancing, and/or covariates pre-specified as the strongest potential confounders from among the original list of covariates. The suggested order of the strongest potential confounder covariates for inclusion in the end point model was: primary refractory, refractory to last treatment, prior ASCT, number of prior therapies, age, LDH, Ann Arbor stage, neutropenia, and anemia.

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. Patients with a starting dosage of lenalidomide of 20 or 25 mg/day, available baseline and at least 1 post-baseline (after the index date) scan, 6 months of follow-up time, physician recorded tumour response assessment, and relevant available clinical data qualified for this validation of response assessment.

Determination of Sample Size

As 81 patients were enrolled and treated in the L-MIND study, the ePS-based 1:1 matching could result in a maximum sample size of n = 162. With an assumed difference of 23% in ORR for lenalidomide monotherapy (35%) versus tafasitamab plus lenalidomide therapy (58%), the achieved power was 80% and minimal detectable difference was 17% in ORR under Fisher’s exact test for unpaired data. Based on an assumed difference of 35% in ORR for lenalidomide monotherapy, ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. Approximately 500 patients needed to be included in the observational cohort to complete successful cohort balancing for 9 dichotomized baseline covariates before conducting comparative efficacy outcome analyses.

Analysis of Outcomes

For the analysis of ORR and CRR, Fisher’s exact tests were used and P values were presented. The primary analysis was based on Fisher’s exact test on the primary analysis population. ORs and 95% CIs were estimated using logistic regression analysis. The cohort status covariate was used in the model. Difference in ORR and CRR between the 2 cohorts was estimated, and exact 95% CI were presented. The number and percentage of patients with CR, PR, SD, PD as best response, and deaths before any post-baseline assessments were presented.

For the analysis of PFS, EFS, OS, and TTNT, the difference in the specific time-to-event measures between the 2 cohorts was compared using the log-rank test. HRs and 95% CI were estimated using Cox proportional hazards model, with cohort status as covariate. For the analysis of DOR, the difference between the 2 cohorts was compared using the log-rank test. Patients without a response were assigned a duration of zero.

Analysis Sets

The primary analysis population was the matched analysis set 25 (MAS 25), which consisted of 1:1 matched patients from the L-MIND study and the observational cohort with a lenalidomide starting dosage of 25 mg per day using 9 baseline covariates.

RE-MIND2

Data such as baseline characteristics, effectiveness outcomes, and treatment termination or dropout due to AEs were retrospectively collected from existing health records, including paper or electronic records of patients treated for R/R DLBCL. Since RE-MIND2 was an observational retrospective study, no patient visits or laboratory tests were required. Only data that had been previously collected were in scope.

Following data collection, the efficacy outcomes of the L-MIND cohort (tafasitamab plus lenalidomide) were compared with the effectiveness in a matched patient population treated with systemic regimens administered in routine clinical care and listed in NCCN or ESMO guidelines^19,20 or pre-specified treatments after cohort balancing using matching and weighting applications of the ePS. Patients who received at least 2 therapy lines for DLBCL were assigned an index date (index date second, third, or fourth line) for each eligible therapy line.

||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Patients were assigned an index date based on the first documented treatment record of the systemically administered therapy for R/R DLBCL under consideration. The pre-index period for each patient was defined as the time between first documented DLBCL diagnosis, or history of cancer other than DLBCL, and the index date (start of R/R DLBCL treatment of second, third, or fourth line). If a patient had received more than 1 treatment regimen (therapy lines) for R/R DLBCL, the patient was assigned an index date for each applicable therapy line. For observational cohorts, the index date for a given line was the start date of any component of R/R DLBCL treatment in that line. Patients who received at least 2 therapy lines for DLBCL were assigned an index date (index date second, third, or fourth line) for the respective therapy line. For the L-MIND study, the index date was the date of first dose of any study drug (lenalidomide or tafasitamab).

The observational period was defined as the time between the index date and end of follow-up. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| A 6-month follow-up rule was implemented in RE-MIND2 as well.

Data were collected for patients who were initially diagnosed with DLBCL between 2010 and 2020. Data from the L-MIND study database, with a data cut-off date of November 30, 2019 (i.e., approximately 2 years after the last patient was enrolled in the study), were used for comparison with the observational cohort of RE-MIND2. For patients in the observational cohort, an analysis window was applied in place of data cut-off date. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The analysis window was applied per therapy line.

Statistical Analysis

To balance the L-MIND cohort with systemically administered therapies, subgroup strata were categorized on the basis of number of lines of therapy (i.e., 2, 3, or 4 therapy lines). Then 1:1 NN matching without replacement was performed using the remaining 8 baseline covariates per stratum to obtain each matched population set. The ratio of the L-MIND cohort to the observational cohort was 1:1, with a maximum ratio of |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The matched population with SMD of 0.2 or less for all baseline characteristics and the highest matching ratio was selected as the primary analysis set for end point calculation. |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| The final matched population for analysis was the aggregation of the matched population of each stratum. Additional matched cohorts were created on the basis of the following 2 subgroups of the L-MIND cohort: (1) 1 prior line before tafasitamab plus lenalidomide, and (2) 2 or 3 prior lines before tafasitamab plus lenalidomide.

To balance the L-MIND cohort with pre-specified treatment regimens, 1:1 NN matching for 9 baseline characteristics was performed using the same covariates as in RE-MIND. Comparative analysis with the L-MIND cohort was performed only if a certain balance of baseline characteristics had been achieved (SMD ≤ 0.2 for all covariates). Patients with different treatment regimens were used in matched population sets under different treatment regimens. Matching was performed only if the number of patients eligible for matching in the pre-specified cohort was larger than number of patients in the L-MIND cohort FAS population used for the RE-MIND2 study (N = 76).

Patients who fulfilled eligibility criteria qualified for matching if they had a baseline response assessment, had a sufficient follow-up for a documented response or progression to the respective treatment regimen, had non–double- or triple-hit genetics lymphoma, were non–transplant-eligible, did not receive ASCT for the given therapy line, had no prior CNS involvement, and had data on all baseline covariates available at the start of the respective treatment.

A pre-specified analysis was not feasible for patients treated with pola-BR and CAR T-cell therapies because of the low number of identified patients. To produce comparative efficacy estimates of the tafasitamab and lenalidomide combination against pola-BR, CAR T-cell therapy, and rituximab and lenalidomide (R2), the following alternative matching methods were conducted in post hoc analyses:

• inverse probability treatment weighting (IPTW) matching using 9 baseline covariates

• IPTW matching using 9 baseline covariates with multiple imputations

• 1:1 NN matching using 6 baseline covariates

• 1:1 NN matching using 6 baseline covariates with multiple imputation

• 1:1 NN matching using 9 baseline covariates with multiple imputation

The 9 covariates used for the post hoc analysis were age, Ann Arbor stage, refractoriness to last therapy line, number of prior lines of therapy, primary refractoriness, prior ASCT, elevated LDH levels, anemia, and neutropenia. For the methods that used only 6 covariates, these were age, refractoriness to last therapy line, number of prior lines of therapy, primary refractoriness, prior ASCT and ECOG status. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Determination of Sample Size

In general, approximately 2,800 patients who had received systemic therapies as per NCCN or ESMO guidelines^19,20 were needed to complete successful cohort balancing before conducting comparative efficacy outcome analyses. The L-MIND primary analysis set consisted of 76 patients for comparison in RE-MIND2. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

Analysis of Outcomes

Time-to-event end points (i.e., OS, PFS, TTNT, DOR, and EFS) were analyzed using standard Kaplan–Meier methodology; log-rank test and HR along with 95% CI and the associated P values estimated using Cox proportional hazard model were reported. Binary end points, including ORR and CR rate, were compared using Fisher’s exact tests, and P values were reported. Treatment effect in terms of difference in ORR or CR rate between the 2 cohorts was estimated, and exact 95% CI was presented. In addition, ORs, ratio of ORR or CR rate, and the ratio of the proportions were presented.

Analysis Sets

The primary analysis in RE-MIND2 used 3 analysis sets: MAS_Pool, MAS_BR, and MAS_R-GemOx. MAS_Pool included patients who met criteria for matching and included 1:1 matched patients from the L-MIND study and the observational cohort using baseline covariates. MAS_BR and MAS_R-GemOx included patients who met criteria for matching using baseline covariates and included 1:1 matched patients from the L-MIND study and observational cohort for each pre-specified treatments.

For the post hoc exploratory analyses of pola-BR and CAR T-cell therapy, the MAS with 9 covariates for specified treatment after multiple imputation (MASMI_9cov) was used to inform the economic model.²⁴ There was a MASMI_9cov for each pre-specified treatment (i.e., MASMI_Pola-BR_9cov, MASMI_CAR-T_9cov).

Results of RE-MIND and RE-MIND2

Summary of Included Studies

RE-MIND

Data from 524 patients were collected from 42 sites in the US and Europe. Following review, data for 34 patients were excluded due to incomplete data, lack of R/R condition, lack of ASCT ineligibility reason, or double- or triple-hit genetics DLBCL. Of the data collected for the 490 patients included, data for 140 patients indicated they had fulfilled the inclusion criteria for RE-MIND, received a lenalidomide starting dose of 25 mg, fulfilled the 6-month follow-up criteria, and had data on the pre-specified baseline covariates available at baseline. Of the 81 patients enrolled in L-MIND, 5 were excluded from the analysis: 1 did not receive lenalidomide, and 4 did not meet the 6-month follow-up criteria. As a result, 76 patients from the L-MIND study were included in RE-MIND.

Following ePS-based NN 1:1 matching, the primary analysis set (MAS 25) comprised 76 patients from each cohort. Patients in the lenalidomide-monotherapy cohort were treated between January 2007 and April 2019, and patients in the combination-therapy cohort received treatment between March 2016 and November 2017. Most patients (63 patients; 82.9%) in the lenalidomide-monotherapy cohort commenced treatment between 2014 and 2019.

Baseline characteristics for the MAS 25 are shown in Table 28 and were generally balanced between the tafasitamab plus lenalidomide combination therapy and lenalidomide-monotherapy cohorts, with SMDs of less than 0.13 for 7 of the 9 baseline characteristics. Residual imbalance was observed for 2 of the 9 covariates: the number of prior lines of therapy (SMD 0.29) and Ann Arbor stage (SMD 0.23). These residual imbalances were addressed in sensitivity analyses that confirmed the primary analysis. There were baseline imbalances for ECOG PS, IPI score, cell of origin, and race, which all had large amounts of missing data. ECOG PS was not a balancing characteristic in the primary analysis but was included as such in a sensitivity analysis.

In the tafasitamab plus lenalidomide and lenalidomide-monotherapy cohorts, the median lenalidomide dose intensity was 17.6 mg per day (interquartile range [IQR] 14.4 to 19.2) versus 19.0 mg per day (IQR 17.5 to 19.5), median follow-up for OS was 21.5 months (IQR 15.1 to 26.5) versus 20.9 months (IQR 15.5 to 29.6), and median time to first post-baseline assessment was 1.9 versus 3.1 months, respectively. In the tafasitamab plus lenalidomide cohort, 96% of assessments were made by CT only or PET-CT, compared with 82% in the lenalidomide-monotherapy cohort; the median frequency of assessment for response in the combination-therapy cohort was 2.1 months (IQR 1.8 to 2.8) and in the lenalidomide-monotherapy cohort, 3.2 months (IQR 1.9 to 4.5).

Table 28: RE-MIND Demographics and Baseline Characteristics – MAS 25

ASCT = autologous stem cell transplant; DLBCL = diffuse large B-cell lymphoma; ECOG PS = Eastern Cooperative Oncology Group performance status; GCB = germinal centre B-cell like; IHC = immunohistochemistry; IPI = International Prognostic Index; LDH = lactate dehydrogenase; MAS = matched analysis set; SD = standard deviation; Q1 = first quartile; Q3 = third quartile; ULN = upper limit of normal.

Source: RE-MIND Clinical Study Report.¹⁶

RE-MIND2

A total of 3,454 patients were enrolled in the observational cohort for RE-MIND2. Of these, 3,449, 1,837, and 894 patients were treated in second, third, and fourth line of anti-DLBCL therapy, respectively. The number of patients eligible for matching in the cohort of systemic therapies pooled was 961. In the cohorts with the pre-specified treatments BR, R-GemOx, CAR T, and pola-BR, there were 282, 235, 51, and 36 patients, respectively. In the 2 cohorts with the pre-specified treatments BR (N = 282) and R-GemOx (N = 235), the number of patients eligible for matching was larger than the number of patients in the L-MIND study FAS population (N = 76). In the L-MIND cohort, 76 of the 81 patients enrolled were eligible for matching (2 patients were excluded because they did not meet the eligibility criteria and 3 patients were excluded because they did not meet the 6 months follow-up rule). Matching exercises and comparative analyses were not performed in other pre-specified treatment cohorts because of the limited number of patients eligible for matching (i.e., less than 76 patients).

Table 29: RE-MIND2 Demographics and Baseline Characteristics – Primary Analysis Sets (MAS_Pool, MAS_BR, MAS_R-GemOx)

ASCT = autologous stem cell transplant; BR = bendamustine plus rituximab; ECOG PS = Eastern Cooperative Oncology Group performance status; IPI = International Prognostic Index; LDH = lactate dehydrogenase; LEN = lenalidomide; MAS = matched analysis set; Pool = systemic therapies pooled; R-GemOx = rituximab plus gemcitabine plus oxaliplatin; SD = standard deviation; Tafa = tafasitamab.

Source: RE-MIND2 Clinical Study Report.¹⁸

Table 30: Redacted