Reimbursement Review

Belantamab Mafodotin (Blenrep), Bortezomib, Dexamethasone

Sponsor: GlaxoSmithKline Inc.

Therapeutic area: Previously treated multiple myeloma

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Abbreviations

Executive Summary

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

Table 1: Background Information on the Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Multiple myeloma (MM) is a progressive and incurable malignancy characterized by the clonal expansion of malignant plasma cells and the overproduction of M proteins.¹ Older individuals are more likely to develop MM, with a higher incidence in men compared to women; and in the US, it is more prevalent in African-American populations relative to people who are white and people of Asian ethnicity.^2,3 The incidence rate of MM increased in Canada from 2003 to 2019 by about 1.3% per year for females and 1.7% per year for males.⁴ In Canada, approximately 3,900 new MM cases were diagnosed in 2023 and there were an estimated 1,700 MM-related deaths.⁵ In 2024, around 4,100 new cases were projected, with 1,750 cases in females and 2,400 in males.⁶ The 5-year survival rate for patients with MM is approximately 50%,⁴ though recent therapeutic advancements have improved survival outcomes, despite the lack of a cure.^7,8

MM is most commonly diagnosed when patients present with symptoms such as bone pain, fatigue, anemia, kidney dysfunction, and recurrent infections.⁹ MM diagnosis requires the presence of 1 or more myeloma-defining events, along with either 10% or more clonal plasma cells in the bone marrow or a biopsy-proven plasmacytoma.³ Myeloma-defining events include end-organ damage (the CRAB criteria of hypercalcemia, renal insufficiency, anemia, and bone lesions) and 3 biomarkers: a clonal bone marrow plasma cell percentage of at least 60%, a free light chain (FLC) ratio of at least 100, and at least 1 focal lesion on an MRI.³ Prognosis is heavily influenced by staging, with the International Staging System (ISS) and its revised version, the Revised International Staging System (R-ISS), widely used to assess disease progression.^10,11 ISS evaluates albumin and B2M levels, where advanced stages correlate with low albumin and high B2M. The R-ISS further incorporates tumour burden and high-risk genetic abnormalities to refine prognostication.^10,11 Other factors, such as age, kidney function, and overall performance status, also impact prognosis.¹² The build-up of resistance to different classes of therapies represents a significant challenge in the clinical management of MM because the disease typically becomes more resistant to treatment with each subsequent line of therapy.^13-15 In Canada, studies show that remission rates decrease with each line of therapy, and attrition rates increase.¹⁶ These findings underscore the progressive nature of the disease and the growing need for diverse therapeutic options that can effectively manage relapsed or refractory multiple myeloma (RRMM). Patients with RRMM often experience a persistent symptom burden, including fatigue, bone pain, and depression, which can severely affect their quality of life (QoL).^17,18 The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of belantamab mafodotin (hereafter referred to as belantamab) in combination with bortezomib and dexamethasone (BVd), administered intravenously once every 3 weeks, at a dose of 2.5 mg/kg, in the treatment of RRMM in adults who have received at least 1 prior therapy.

Perspectives of Patient, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups that responded to the CDA-AMC call for input and from clinical experts consulted by CDA-AMC for the purpose of this review.

Patient Input

Patient input was submitted by 1 patient group: Myeloma Canada. Information was gathered from patients and caregivers through an online survey (N = 292). The most important symptoms related to myeloma that respondents want to control, in order of importance, include infections, mobility, renal problems, and pain. About 41% of patient respondents indicated that they required the help of a caregiver to manage their disease or treatment-related symptoms. The input noted that patients’ disease symptoms impacted their ability to travel, work, exercise, and conduct volunteer activities. Respondents noted that the most significant financial implications due to myeloma treatment was the loss of income and/or pension funds due to absence from work, disability, or early retirement, followed by the costs of travel, parking, drugs, and accommodation. Respondents felt that the interruption of life goals or accomplishments had the greatest impact on their QoL, followed by the loss of sexual desire, and anxiety or worry. Respondents reported a desire for a treatment that extends their life expectancy without disease progression (67% of respondents rated this “extremely important”) and improves their QoL (58% rated this “extremely important”).

No survey respondents had experience with BVd and 10 respondents indicated they had experience with belantamab (7 respondents with belantamab plus pomalidomide and dexamethasone and 3 respondents either with belantamab as monotherapy or combined with dexamethasone). Among these respondents, the least bearable side effects reported included blurry vision, dry eyes, eye irritation, sensitivity to light, and infections. Eight of the respondents indicated that the overall side effects of belantamab treatment were “somewhat or mostly” manageable. When asked if the treatment was effective in controlling myeloma, 6 respondents reported “mostly or completely,” 3 respondents stated “somewhat,” and 1 respondent stated “slightly.” The input emphasized the importance of patient preference in weighing the potential costs and benefits of any new treatment.

Clinician Input

Input From Clinical Experts Consulted for This Review

The clinical experts consulted by Canada’s Drug Agency (CDA-AMC) highlighted several challenges associated with the currently available therapies for MM in Canada. First, while various classes of drugs are available to manage MM at different stages, no single treatment can offer a definitive cure for the disease. Second, a critical treatment gap exists for patients who become refractory to lenalidomide or anti-CD38 therapies. This is especially problematic for older adults who have previously received daratumumab plus lenalidomide and dexamethasone because limited therapeutic options are available upon relapse. Third, treatment regimens such as carfilzomib plus dexamethasone (Kd) or selinexor plus bortezomib plus dexamethasone (SVd) demonstrate suboptimal outcomes, and many patients may not be candidates for more aggressive therapies like chimeric antigen receptor T-cell (CAR T-cell) therapy or bispecific antibodies. Additionally, there is a need for therapies with improved toxicity profiles. Because patients with MM often require treatment regimens with manageable side effects to preserve their QoL, there is an increasing demand for therapies that involve fewer clinic visits and less intensive monitoring.

According to the clinical experts consulted, belantamab would primarily be used in second-line treatment and beyond for MM, specifically for patients who have relapsed or are refractory to prior therapies. The clinical experts indicated that older patients, particularly those who have previously received lenalidomide plus bortezomib and dexamethasone, or daratumumab in combination with lenalidomide and dexamethasone, and are not refractory to bortezomib, are ideal candidates for belantamab. They also emphasized that, aside from ocular toxicities, the toxicity profile of belantamab is relatively favourable, particularly in older patients. Unlike other therapies, belantamab does not exhibit significant myelosuppressive toxicities, making it a good option for older patients who are more vulnerable to the adverse effects of other treatments. However, the clinical experts indicated that ocular toxicities remain a concern, and patients with pre-existing ocular conditions would be least suitable for belantamab.

The clinical experts indicated that assessing response to treatment for MM is generally straightforward, with routine blood tests being the primary method of monitoring. According to clinical experts, ocular toxicity with belantamab therapy can negatively impact QoL and requires more intensive monitoring than is standard in care for MM. Specifically, regular ophthalmologic assessments and slit lamp examinations are necessary to guide dose adjustments and prevent more severe ocular toxicity because they are not typically part of standard clinical practice in MM. The clinical experts consulted indicated that treatment with BVd would be discontinued for 2 primary reasons: disease progression and intolerable toxicity, with ocular toxicity being the most significant. Furthermore, the clinical experts also noted that ocular toxicity may require extending dosing intervals to every 8 weeks to 12 weeks, and in severe or intolerable cases, treatment may be discontinued. According to the clinical experts, treatment of MM with belantamab should be managed by trained hematologist-oncologists, with eye specialists involved in monitoring and managing ocular toxicity.

Clinician Group Input

Two clinician groups consisting of a total of 27 clinicians provided input for this review: the Ontario Health (Cancer Care Ontario) (OH [CCO]) Hematology Cancer Drug Advisory Committee and the Canadian Myeloma Research Group (CMRG). Both submissions noted that myeloma remains incurable despite the introduction of new drugs over the last 2 decades. Patients eventually become refractory to all available funded antimyeloma drugs. The input stated that symptom burden for patients with MM is high, with patients experiencing bone pain and destruction, anemia and other cytopenias, renal damage, hypercalcemia, and a high risk of infection. The CMRG stated that despite the clear benefits of lenalidomide as part of first-line therapy, progression on this potent drug even as single-drug maintenance leads to shorter progression-free survival (PFS) outcomes with nearly all traditional and reimbursed second-line regimens (including those containing an anti-CD38 monoclonal antibody) compared to the results without such exposure. Therefore, the input noted that drug exposure, rather than lines of therapy, more accurately defines the need for access to innovative treatments to forestall the development of refractory myeloma. The CMRG also noted that with the movement of combinations of 3 major drug classes (i.e., an immunomodulatory drug [IMiD], a proteasome inhibitor, and an anti-CD38 monoclonal antibody) to the first-line and second-line treatment setting, exposure and resistance to multiple drug classes now occur much earlier in the disease course than in the past. Therefore, the highest unmet need in myeloma continues to be effective treatment for patients who have progressed despite exposure to effective drugs.

OH (CCO) noted that the regimen would be another second-line treatment option for patients who are sensitive to bortezomib. The input noted that the treatment under review would be most suitable for patients who are unlikely to receive CAR T-cell therapy because BVd may preclude the future use of B-cell maturation antigen (BCMA)–targeted CAR T-cell therapy. Standard myeloma response outcomes used in clinical practice would be appropriate to determine if a patient is responding to the treatment under review, based on the M protein markers in the serum and/or urine, a bone marrow biopsy and, in some instances, imaging studies. Clinically meaningful responses usually correlate with at least a partial remission as defined by the International Myeloma Working Group (IMWG) consensus criteria. These include improvement in symptoms (the cessation of bone destruction with less pain, fewer fractures, and less need for radiotherapy), improvement in energy, and a better ability to perform activities of daily living. Responses are generally assessed every 1 month to 3 months depending on clinical stability and the regimen used for therapy. Factors to consider when deciding to discontinue the treatment under review include significant toxicities, particularly ocular adverse events (AEs), as well as disease progression. The input stated that the appropriate setting for BVd treatment is an outpatient setting and there is also a need for ophthalmological assessments.

Drug Program Input

Input was obtained from the drug programs that participate in the CDA-AMC reimbursement review process. The following were identified as key factors that could potentially affect the implementation of a CDA-AMC recommendation for BVd:

• relevant comparators

• considerations for the prescribing of therapy

• generalizability

• care provision issues

• system and economic issues.

The clinical experts consulted by CDA-AMC provided advice on the potential implementation issues raised by the drug programs. Refer to Table 4 for more details.

Clinical Evidence

Systematic Review

Description of Studies

The sponsor-conducted systematic literature review (SLR) identified 1 pivotal, open-label, randomized trial (the DREAMM-7 study, N = 494) that assessed the efficacy and safety of BVd relative to daratumumab in combination with bortezomib and dexamethasone (DVd) in adult patients with RRMM. The primary objective of the DREAMM-7 study was to demonstrate the superiority of BVd compared to DVd in PFS. Key secondary objectives were to demonstrate superiority in overall survival (OS), duration of response (DoR), and minimal residual disease (MRD) negativity. The trial enrolled patients who had received at least 1 prior line of MM therapy and had documented disease progression during or after their most recent treatment. Patients were enrolled in the trial across 142 centres in 20 countries, including 5 sites in Canada. In the DREAMM-7 trial, patients were randomized in a 1:1 ratio to receive BVd at the dose of 2.5 mg/kg on day 1 of every 21-day cycle intravenously, or DVd. The DREAMM-7 study included a screening period, a treatment period, and a follow-up period. Efficacy end points of interest to this review included PFS, OS, health-related quality of life (HRQoL) measured by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) global health status (GHS), DoR, MRD negativity, overall response rate (ORR), complete response rate (CRR), and safety outcomes, including ocular AEs and serious adverse events (SAEs). The trial end points were analyzed using data from the cut-off date of October 2, 2023, and the database lock date of November 6, 2023, for the first interim analysis or interim analysis 1 (IA1), and from the cut-off date of October 7, 2024, for the second interim analysis or interim analysis 2 (IA2) or a prespecified primary PFS analysis.

In general, patient demographic and disease characteristics were well balanced between the BVd and DVd treatment groups. The mean age of patients was 64.5 (standard deviation [SD] = 9.5) years in the BVd group and 63.6 (SD = 10.1) years in the DVd group. Most patients had an Eastern Cooperative Oncology Group Performance Status (ECOG PS) score of 0 or 1 (96% in both groups), and an R-ISS score of stage I or stage II (95% and 94% in the BVd and DVd groups, respectively), with immunoglobulin G being the most common myeloma immunoglobulin. Of the 494 patients, 250 (51%) patients had received 1 prior line of therapy, 117 (24%) patients had received 2 prior lines of therapy, and 70 (14%) patients had received 3 prior lines of therapy. Additionally, 257 (52%) patients had prior exposure to lenalidomide, and 166 (34%) patients had disease refractory to lenalidomide. High-risk cytogenetic abnormalities were present in 67 (28%) patients in the BVd group and 69 (27%) patients in the DVd group, with t(4;14) and 17p13del being the most common high-risk cytogenetic abnormalities in both groups.

Efficacy Results

PFS per Independent Review Committee Assessment

At the time of IA1, using the October 2, 2023, data cut-off, the median duration of follow-up was 28.2 months. In the intention-to-treat (ITT) population, PFS events were reported for 91 (37%) patients in the BVd group and 158 (63%) patients in the DVd group. The median PFS was 36.6 months (95% confidence interval [CI], 28.4 months to not estimable) in the BVd group and 13.4 months (95% CI, 11.1 months to 17.5 months) in the DVd group, with a between-group hazard ratio (HR) of 0.41 (95% CI, 0.31 to 0.53; P < 0.00001) in favour of the BVd group. The Kaplan-Meier (KM)–estimated probabilities of being alive or progression-free at 12 months and 18 months were 78.3% (95% CI, 72.2% to 83.2%) versus 53.3% (95% CI, 46.5% to 59.5%), and 68.8% (95% CI, 62.0% to 74.7%) versus 42.8% (95% CI, 36.2% to 49.2%), in the BVd and DVd groups, respectively.

PFS reached the predefined boundary for statistical significance at IA1 and, in accordance with the study protocol, was evaluated as an exploratory end point at the time of IA2. At the time of IA2, using the October 7, 2024, data cut-off, the median PFS was ████ ██████ ████ ███ ████ ██████ ██ ████ ███████ in the BVd group and ████ ██████ ████ ███ ████ ██████ ██ ████ ███████ in the DVd group. The KM-estimated probabilities of being alive and progression-free at 24 months and 36 months were █████ ████ ███ █████ ██ ██████ versus █████ ████ ███ █████ ██ █████), and █████ ████ ███ █████ ██ ██████ versus █████ ████ ███ █████ ██ ██████, in the BVd and DVd groups, respectively.

The results of predefined exploratory subgroup analyses were consistent with the results of the primary PFS analyses, including for those exposed or refractory to lenalidomide (HR = 0.33 [95% CI, 0.23 to 0.48] and HR = 0.37 [95% CI, 0.24 to 0.56], respectively), those with more than 1 prior line of therapy (HR = 0.33 [95% CI, 0.23 to 0.48]), those with high cytogenetic risk (HR = 0.36 [95% CI, 0.22 to 0.58]) and those with R-ISS stage II or stage III (HR = 0.45 [95% CI, 0.32 to 0.64]). The results of all sensitivity analyses were consistent with the primary PFS analyses. The results of the PFS analyses per an independent review committee (IRC) assessment were consistent with those from the investigator assessment.

OS per IRC Assessment

At the time of IA1, using the October 2, 2023, data cut-off, the median OS was not reached in either treatment group. OS data had reached 29% (141 of 494) of patients for overall maturity. Death was reported in 54 (22%) patients in the BVd group and 87 (35%) patients in the DVd group, with a between-group HR of 0.57 (95% CI, 0.40 to 0.80; P value = 0.00049). At the time of IA1, the results for OS did not meet the significance criterion.

At the time of IA2, using the October 7, 2024, data cut-off, the median OS had not been reached in either treatment group. Death was reported in 68 (28%) patients in the BVd group and 103 (41%) patients in the DVd group, with a between-group HR of 0.58 (95% CI, 0.43 to 0.79; P value = 0.00023) in favour of the BVd group. At the time of IA2, OS data had reached 34.6% (171 of 494) of patients for overall maturity, and the significance criterion had been met (P < 0.00112). The KM-estimated probabilities of being alive at 24 months and 36 months were 79.0% (95% CI, 73.2% to 83.7%) versus 67.4% (95% CI, 61.0% to 73.0%), and 71.4% (95% CI, 67.9% to 79.2%) versus 60.2% (95% CI, 53.6% to 66.2%), in the BVd and DVd groups, respectively.

Duration of Response

DoR was analyzed based on the restricted mean duration of response (RMDoR), using a nonparametric approach. The RMDoR is a composite end point that integrates overall response data and PFS data, allowing cross-group comparison, and efficiently assesses the treatment effect related to tumour reductions. At the time of IA1, using the October 2, 2023, data cut-off, the RMDoR was 19.0 months (95% CI, 17.7 months to 20.4 months) in the BVd group and 13.2 months (95% CI, 11.8 months to 14.6 months) in the DVd group, with a ratio of RMDoR of ████ ████ ███ ████ ██ █████ P value < 0.00001) in favour of the BVd group.

At the time of IA1, using the October 2, 2023, data cut-off, the median DoR was 35.6 months (95% CI, 30.5 months to not estimable) in the BVD group and 17.8 months (95% CI, 13.8 months to 23.6 months) in the DVd group. At the time of IA2, using the October 7, 2024, data cut-off, the median DoR was 40.8 months (95% CI, 30.5 months to not estimable) in the BVd group and 17.8 months (95% CI, 13.8 months to 23.6 months) in the DVd group. The KM-estimated probabilities of maintaining response at 12 months and 36 months were 83.3% (95% CI, 77.1% to 87.9%) versus 61.2% (95% CI, 53.4% to 68.0%), and █████ ████ ███ █████ ██ ██████ versus █████ ████ ███ █████ ██ ██████, in the BVd and DVd groups, respectively.

Minimal Residual Disease

MRD is considered to be a biomarker that provides a reliable quantification of tumour burden, independent of the assay used.¹⁹ At the time of IA1, using the October 2, 2023, data cut-off, the MRD negativity analysis was considered exploratory because the OS results were not significant at the time of IA1. The proportion of patients who had achieved MRD negativity by best response (complete response [CR] or stringent complete response [sCR]) was higher in the BVd group compared with the DVd group at 24.7% versus 9.6%, respectively (P < 0.00001). Because the IA2 analysis of OS reached statistical significance, the MRD negativity results from the IA1 primary analysis can be considered statistically significant due to the prespecified hierarchical testing of outcomes. At the time of IA2, using the October 7, 2024, data cut-off, MRD negativity rates by CR or sCR remained consistent with the IA1 results. The BVd group showed higher MRD negativity rates (25.1%) compared with the DVd group (10.4%).

Complete Response Rate

At the time of IA1, using the October 2, 2023, data cut-off, a greater proportion of patients in the BVd group (34.6%) compared with the DVd group (17.1%) had achieved CR or sCR, with a between-group difference of █████ ████ ███ ████ ██ ███████

Overall Response Rate

At the time of IA1, using the October 2, 2023, data cut-off, a greater proportion of patients in the BVd group (82.7%) compared with the DVd group (71.3%) had achieved a confirmed partial response (PR) or better, with a between-group difference of █████ ████ ███ ████ ██ ███████

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

GHS Domain

In the DREAMM-7 study, HRQoL was assessed using the EORTC QLQ-C30 tool. EORTC QLQ-C30 data reported in this section are from IA1, with the data cut-off date of October 2, 2023. The EORTC QLQ-C30 GHS domain remained stable across both treatment groups over time. Both groups showed a slight mean deterioration in GHS scores between week 4 and week 43, after which scores stabilized and remained consistent throughout the treatment period. Between week 43 and week 100, 24% to 35% of patients in the BVd group experienced an improvement in GHS or QoL scores (a 10-point increase or higher). Small sample sizes at later time points limited the ability to interpret the results.

At week 43, in the GHS score, patients in the BVd group experienced a mean change of ███ ██████ from baseline compared with a change of ███ ██████ in the DVd group, with a least squares mean difference between groups of ███ ████ ███ ████ ██ ████. At week 121, in the GHS score, patients in the BVd group experienced a mean change of ███ ██████ from baseline compared with a change of ████ ██████ in the DVd group, with a least squares mean difference between groups ███ ████ ███ ████ ██ █████

Harms Results

In the DREAMM-7 trial, all patients in both groups experienced at least 1 AE. At the time of IA1, a higher proportion of patients in the BVd group had experienced at least 1 SAE compared to those in the DVd group (50% versus 37%, respectively). The most common SAEs in the BVd and DVd groups were pneumonia (11% versus 4%, respectively), COVID-19 (5% versus 4%, respectively), pyrexia (5% versus 4%, respectively), and COVID-19 pneumonia (93% in both groups). The IA2 data on SAEs were not available at the time of submission to CDA-AMC. At the time of IA1 and IA2, a higher proportion of patients in the BVd group had discontinued treatment due to AEs compared to the DVd group (31% versus 19%, and 32% versus 19%, respectively). At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced a dose interruption or delay of study treatment compared to the DVd group (94% versus 75%, and 95% versus 76%, respectively). At the time of IA2, deaths were reported in 29% of patients in the BVd group and in 41% of patients in the DVd group. The causes of death were similar between groups, with the exception of those attributed to cancer. The majority of deaths in both groups were attributed to cancer, with 10% in the BVd group and 22% in the DVd group —primarily due to MM.

Notable Harms

At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced at least 1 ocular AE of any grade compared to the DVd group (79% versus 29%, and ███ ██████ ████ ████████████). Blurred vision and dry eye were reported in more than half of the patients in the BVD treatment group. Ocular AEs in the BVd group were more severe, with ███ classified as grade 3 or grade 4, while the majority in the DVd group were classified as grade 1 █████. At the time of IA2, treatment discontinuations due to ocular AEs were reported in ██ of patients in the BVd group compared to none in the DVd group. Treatment modifications due to ocular AEs were more common in the BVd group, with 52% of patients experiencing ocular AEs having doses interrupted, delayed, or reduced compared to only 2% in the DVd group. At the time of IA1, investigator-assessed corneal events and best corrected visual acuity (BCVA) events, as evaluated by the Keratopathy and Visual Acuity (KVA) scale, were reported exclusively in patients in the BVd group. A total of 201 (83%) patients in the BVd group had an incidence of corneal events (overall KVA grade). The severity of the corneal events was high, with ███ of the 201 patients classified as grade 3 and ███ as grade 4. Most corneal events were managed through dose modifications, with 89% of affected patients having doses interrupted or delayed, and 31% experiencing dose reductions. In the BVd group, 81% of patients had BCVA events, mostly classified as grade 2 or higher. The incidence of investigator-assessed BCVA events, as evaluated by the KVA scale, was ███ in the BVd group, with the majority of events classified as grade 2.

At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced thrombocytopenia compared to those in the DVd group (87% versus 65%, and 88% versus 65%, respectively). At the time of IA2, of the 212 patients in the BVd group and 160 patients in the DVd group who had experienced thrombocytopenia, grade 3 and grade 4 thrombocytopenia were more frequent in the BVd group (83%) compared to the DVd group (71%). Thrombocytopenia events of grade 4 were reported in 53% of patients receiving BVd versus 33% of patients receiving DVd. In the DREAMM-7 trial, ███ of patients in the BVd group and ██ in the DVd group had received platelet transfusions. At the time of IA2, despite the high incidence, few patients (5% or less) in either group had experienced SAEs related to thrombocytopenia, and treatment discontinuation rates due to thrombocytopenia were low (██ in the BVd group and ██ in the DVD group). Dose reductions and interruptions or delays were more common in the BVd group compared to the DVd group (43% versus 20%, and 53% versus 39%, respectively). At the time of IA1, neutropenia was reported in ███ of patients in the BVd group and ███ of patients in the DVd group. Neutropenia events of grade 4 were more common in the BVd group (███) compared to the DVd group (███). A higher proportion of patients in the DVd group had experienced infusion-related reactions (███) compared to the BVd group (██), with more than ███ of patients in both treatment groups having grade 2 events.

Critical Appraisal

Randomization was performed using an appropriate methodology with adequate allocation concealment; randomization stratification was prespecified. Knowledge of the assigned treatment due to open-label design of the trial could have led to bias in the reporting and measurement of subjective outcomes, including patient-reported outcomes (e.g., HRQoL) and subjective AEs. However, the extent and direction of bias due to treatment knowledge is uncertain. Based on the enrolled sample size, the study was powered to test its primary and key secondary end points. The statistical analysis methods appear to be acceptable. Results for the primary outcome were based on a prespecified interim analysis (249 PFS events), which occurred at an 89% information fraction, relative to the planned final PFS analysis (280 PFS events). The OS data were immature at both interim analyses and the median OS was not reached in either treatment group. Because the study outcomes were based on interim analyses, there is a risk that the effect of BVd compared with DVd was overestimated; however, the existence and extent of any overestimation remains uncertain.^20-22 Patients were allowed to receive posttreatment anticancer medications after discontinuing study treatment, which could have influenced the assessment of OS. Subgroup analyses were prespecified and conducted only for PFS; however, there was no multiplicity control for the subgroup analyses. A large number of patients in the DREAMM-7 trial discontinued treatment, with fewer discontinuations (both from monotherapy and combination therapy) in the BVd group (66%) compared to the DVd group (78%), which may influence the interpretation of harms outcomes. The clinical experts consulted indicated that the main reasons for treatment discontinuation were as expected, and included disease progression and AEs, with ocular toxicity being the main concern associated with belantamab treatment. They also noted that several strategies have been developed to manage the ocular toxicity of belantamab and prevent permanent drug discontinuation. No strong conclusions could be drawn about the effect of BVd compared with DVd on HRQoL due to an increased risk of type I error and a high risk of attrition bias, particularly at longer follow-up.

According to the clinical experts consulted by CDA-AMC for this review, the patient population in the DREAMM-7 trial generally reflects patients in clinical practice in this setting. Patients were eligible for inclusion in the DREAMM-7 trial if they met 1 of the measurability parameters, including having a serum M protein concentration of 0.5 g/dL or higher. The clinical experts noted that some patients with serum M protein concentration below 0.5 g/dL would still be eligible for treatment with BVd in clinical practice. The clinical experts indicated that, aside from refractoriness to daratumumab as a comparator in this trial, all other exclusion criteria were consistent with clinical practice. According to the clinical experts consulted, the demographic and disease characteristics of the DREAMM-7 trial population were reflective of patients living in Canada with RRMM. The mean age of patients in the DREAMM-7 study was 64 years, with clinical experts noting that in the real-world setting, transplant-eligible patients would likely be younger, whereas patients with relapsed or refractory disease would typically be older. Most patients in the DREAMM-7 trial had an ECOG PS score of 0 or 1, and an R-ISS score of stage I or stage II. The clinical experts consulted indicated that this would be reflective of clinical practice in the second-line setting, but not in later relapse. Only 34% of patients in the trial were refractory to lenalidomide. The clinical experts indicated that, in the real-world setting, the proportion of patients who are refractory to lenalidomide in second-line therapy and beyond would be higher because it is commonly used as maintenance therapy after achieving a response to initial treatment. They also noted that these patients may have a poorer response to second-line or third-line treatments compared to those who are not refractory to lenalidomide.

GRADE Summary of Findings and Certainty of the Evidence

For pivotal studies and randomized controlled trials (RCTs) identified in the sponsor’s systematic review, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool was used to assess the certainty of the evidence for outcomes considered most relevant to inform expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group^23,24

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The selection of outcomes for GRADE assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members:

• PFS (the time from the date of randomization until the earliest date of documented disease progression determined by an IRC, according to IMWG criteria,²⁵ or death due to any cause)

• OS (the time from the date of randomization until the date of death due to any cause)

• HRQoL (the change from baseline in GHS or QoL scale scores)

• harms (severe AEs and ocular AEs).

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for BVd.

Table 2: Summary of Findings for BVd Versus DVd for Patients With RRMM

BVd = belantamab mafodotin plus bortezomib and dexamethasone; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; DVd = daratumumab plus bortezomib and dexamethasone; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; GHS = global health status; HRQoL = health-related quality of life; IA1 = interim analysis 1; IA2 = interim analysis 2; KM = Kaplan-Meier; QoL = quality of life; MID = minimal important difference; NA = not applicable; NR = not reported; OS = overall survival; PFS = progression-free survival; RCT = randomized controlled trial; RRMM = relapsed or refractory multiple myeloma.

Notes: Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the Table 2 footnotes.

Estimates of the absolute between-group differences were calculated using the KM method at the request of CDA-AMC.

^aIA2, using the October 7, 2024, data cut-off.

^bPFS reached the predefined boundary for statistical significance at IA1 and, in accordance with the study protocol, was evaluated as an exploratory end point at the time of IA2.

^cAn empirically derived MID was not identified for the between-group difference for this outcome. A difference of 10% (100 per 1,000 patients) between the groups was identified by the clinical expert consulted by CDA-AMC as a threshold of clinical importance for this outcome.

^dAt the time of IA2, using the October 7, 2024, data cut-off, OS met the significance criterion of P being less than 0.00112.

^eRated down 2 levels for serious study limitations. Results were based on an interim analysis, so there was the potential for effect to be overestimated, and the OS data were immature. The median OS was not reached in either treatment group. There was a risk of bias due to confounding as a result of the transition of patients to subsequent treatment postprogression. There was no established between-group MID for OS and the clinical experts considered that a 5% difference (50 per 1,000 patients) between groups in the probability of patients being alive could be considered a threshold of clinical importance. The point estimate suggests a clinically important benefit while the lower boundary of the 95% CI suggests little to no difference.

^fIA1, using the October 3, 2023, data cut-off.

^gRated down 2 levels for very serious risk of bias due to a relatively high amount of missing data (direction unclear) over time and due to assessor knowledge of treatment assignment. Also rated down 1 level for serious imprecision; CIs were wide, especially at later time points, and included the potential for benefit and harm. Because no single time point was of highest importance, the trend over time was appraised.

^hThe statistical testing for HRQoL was not adjusted for multiplicity in the DREAMM-7 trial and should be considered as supportive evidence.

ⁱThere was no established MID for this outcome and clinical experts consulted by CDA-AMC could not provide a threshold of important difference. In the absence of a known threshold, the null was used.

^jA large number of patients in the DREAMM-7 trial discontinued treatment, with fewer discontinuations (both from monotherapy and combination therapy) in the BVd group (66%) compared to the DVd group (78%); patients in the BVd group had longer exposure to the study drug than those in the DVd group. Rated down 1 level for imprecision; an empirically derived MID was not identified for the between-group difference for this outcome. Based on the MID identified by clinical experts (a difference of 10% between the groups), the 95% CI for the between-group difference crossed the MID threshold.

Sources: Clinical Study Report for the DREAMM-7 study (2024)²⁶ and Clinical Study Report Addendum (Interim Analysis 2) for the DREAMM-7 study (2025).²⁷

Long-Term Extension Studies

No long-term extension studies were identified by the sponsor.

Indirect Comparisons

Description of Studies

In the absence of direct comparative data for belantamab versus comparators used in clinical practice aside from daratumumab, the sponsor supplied 1 report that included indirect treatment comparisons (ITCs). The primary objective of the ITC was to assess the comparative efficacy of BVd in a population matching the ITT population of the DREAMM-7 trial, relative to other comparators for which PFS data were available. The secondary objectives of the NMA were to assess the comparative efficacy of BVd in a population of prior lenalidomide-exposed patients, based on available data for OS and ORR, and in a population of lenalidomide-refractory patients, relative to comparators for which data were available for PFS and ORR. The following outcomes were reported to address the objectives of the network meta-analysis (NMA): PFS, OS, and ORR for the ITT population, and PFS and ORR for the lenalidomide-exposed and lenalidomide-refractory populations.

Efficacy Results

The NMA results presented in this report were limited to comparisons between BVd and regimens relevant to Canada for this submission:

• isatuximab plus high-dose carfilzomib and dexamethasone (IhKd)

• high-dose carfilzomib plus dexamethasone (hKd)

• pomalidomide plus bortezomib and dexamethasone (PVd)

• SVd

• DVd

• bortezomib plus dexamethasone (Vd)

• Kd.

Progression-Free Survival

In the ITT population, results from the fixed-effects models indicated that BVd demonstrated a favourable improvement in PFS compared with all 7 comparators, with HRs ranging from 0.13 (95% credible interval [CrI], 0.09 to 0.18) versus Vd to 0.42 (95% CrI, 0.26 to 0.69) versus IhKd. The results from the corresponding random-effects model showed a favourable improvement in PFS for BVd compared with DVd, hKd, PVd, SVd, and Vd, with HRs ranging from 0.13 (95% CrI, 0.06 to 0.28) versus Vd to 0.41 (95% CrI, 0.22 to 0.75) versus DVd. However, the evidence regarding PFS between BVd and IhKd, as well as Kd, was insufficient to show a difference between groups. The point estimates were comparable across both the fixed-effects and random-effects models.

Among lenalidomide-exposed patients, results from the fixed-effects models found that BVd demonstrated a favourable improvement in PFS compared with all comparators. The results from the corresponding random-effects model showed a favourable improvement in PFS for BVd versus DVd, hKd, and Vd; evidence versus other comparators (IhKd, Kd, PVd, and SVd) were insufficient to show a difference between groups. Among lenalidomide-refractory patients, results from the fixed-effects models found that BVd demonstrated a favourable improvement in PFS compared with all comparators. The results from the corresponding random-effects model showed a favourable improvement in PFS of BVd versus DVd, hKd, and Vd; evidence versus other comparators (IhKd, Kd, and PVd) was insufficient to show a difference between groups.

Overall Survival

In the ITT population, results from the fixed-effects models indicated that BVd demonstrated a favourable improvement in OS compared with DVd, hKd, PVd, SVd, and Vd, with HRs ranging from 0.39 (95% CrI, 0.26 to 0.59) compared to Vd to 0.57 (95% CrI, 0.40 to 0.80) compared to DVd. However, results from the corresponding random-effects model for OS revealed that the evidence for a difference between BVd and other comparators was insufficient to show a difference between groups. No secondary or subgroup analyses for OS were conducted or reported in the sponsor-submitted NMA.

Overall Response Rate

In the ITT population, results from fixed-effects models found that BVd demonstrated a favourable improvement in ORR compared with DVd, hKd, PVd, SVd, and Vd, with ORs ranging from 1.94 (95% CrI, 1.27 to 2.99) compared to DVd to 5.99 (95% CrI, 3.41 to 10.64) compared to Vd. The results from the corresponding random-effects model for PFS showed that BVd demonstrated a favourable improvement in ORR compared with Vd alone; evidence for other comparisons was insufficient to show a difference between groups. The point estimates were comparable across the fixed-effects and random-effects models. Among lenalidomide-exposed patients, results from fixed-effects models found that BVd demonstrated a favourable improvement in ORR compared with DVd, hKd, high-dose carfilzomib plus daratumumab and dexamethasone (hkDd), PVd, SVd, and Vd. The results from the corresponding random-effects model showed a favourable improvement in ORR versus only Vd; evidence for the other comparisons was insufficient to show a difference between groups. Among lenalidomide-refractory patients, results from fixed-effects models found that BVd demonstrated a favourable improvement in ORR compared with SVd and Vd. The results from the corresponding random-effects model showed a favourable improvement in ORR versus Vd alone; evidence for the other comparisons was insufficient to show a difference between groups. No subgroup analyses for ORR were conducted or reported in the sponsor-submitted NMA.

Harms Results

No comparative effect estimates for harms were provided.

Critical Appraisal

Overall, the SLR conducted to identify potentially relevant studies for the ITC was methodologically robust. The clinical experts indicated that the most relevant comparators included in the NMA are IhKd for transplant-eligible patients with MM and SVd for transplant-ineligible patients. The clinical experts also noted that clinicians would prefer prescribing IhKd over hKd, as well as Kd, in transplant-ineligible patients in these settings. Additionally, the clinical experts noted that cyclophosphamide plus bortezomib and dexamethasone (CyVd), cyclophosphamide plus high-dose carfilzomib and dexamethasone (CyhKd), panobinostat plus bortezomib and dexamethasone (PanoVd), hkDd, and elotuzumab plus bortezomib and dexamethasone (EVd), which are included in this NMA, are not relevant to clinical practice in Canada for RRMM. The clinical experts indicated that cyclophosphamide in combination with doublet therapies (Kd and Vd) is a relevant comparator for BVd in the second-line setting and beyond; however, it was not included in the sponsor-submitted NMA. Therefore, the relative effect of BVd versus this comparator of interest is unknown.

Several important sources of heterogeneity were noted across the included trials in terms of study design (e.g., crossover, placebo as a comparator, the inclusion of phase II trials, the time during which trials were undertaken, dosing regimens). Notably, there was considerable variation across studies in the number of prior lines of therapy, prior lenalidomide and IMiD exposures, and high cytogenic risk profile. The observed heterogeneity could have introduced bias that was not adequately accounted for, leading to uncertainty in the analysis. This uncertainty is reflected in the wide CrIs of the comparisons of treatment effects, particularly for ORR and OS. Several important disease-specific characteristics were not reported in some studies within the network, which limits the ability to assess heterogeneity between the studies. While the definitions of end points were similar across the trials, there was considerable variation in the duration of follow-up, and the censoring rules across the PFS analyses were not specified, which may represent a potential source of heterogeneity. The magnitude of the treatment effects appeared consistent across both subgroups (lenalidomide-exposed and lenalidomide-refractory); however, the 95% CrIs were wide given the smaller sample sizes, leading to greater uncertainty in the resulting treatment effects. The studies included in the secondary and subgroup analyses for PFS differ from those in the full network, meaning the subgroup analyses do not directly compare the same network as the full population analyses. Narrowing the network could potentially impact the results. The studies within networks included a mix of patients in multiple lines of therapy, which may have introduced bias because patients in earlier or later lines of therapies can influence each network differently. According to the clinical experts consulted by CDA-AMC, patients in earlier lines of therapy are likely to have better outcomes compared with patients in later lines of therapy. There were no closed loops in the network to allow for consistency to be assessed.

Bayesian fixed methods were used as a base-case analysis, with a random-effects model for the exploratory analysis. However, if heterogeneity is present, the CrIs of the fixed-effects model may be narrower and less conservative, potentially underestimating the uncertainty from between-study variation. The assumptions of the fixed-effects model (e.g., a single true effect size, no between-study heterogeneity) are likely less realistic than those of the random-effects model (e.g., assumes variation in effect sizes across studies). Secondary analyses were performed for lenalidomide-exposed and lenalidomide-refractory patient populations because it is expected that most patients living in Canada will receive a regimen based on lenalidomide in the first line of therapy, with subsequent therapies tailored to the patient’s initial response. Some trials had concerns for risk of bias; this assessment was undertaken at the study level, which ignores that risks of bias may vary by effect estimates. It may not be assured that randomization was maintained within subgroups (unless these were stratification factors in contributing trials). Additionally, no patient-reported QoL data, which was considered an important end point for this review, were evaluated. Furthermore, no comparative effect estimates for harms were provided. These limitations preclude a comprehensive assessment of the balance of benefits and harms and must be considered when drawing conclusions from the NMA results.

Studies Addressing Gaps in the Evidence From the Systematic Review

No studies addressing gaps in the pivotal and RCT evidence were identified by the sponsor.

Conclusions

The available evidence from the pivotal, open-label, multicentre RCT (the DREAMM-7 study) provides important insights into the efficacy and safety of BVd compared with DVd in adults with RRMM who have received at least 1 prior therapy. Compared with DVd, BVd results in a clinically important increase in the probability of being alive and progression-free at 12 months and 36 months. At 36 months, BVd likely results in a clinically important increase in the probability of being alive when compared with DVd. At the time of both interim analyses, median OS had not been reached in either group, and the effects of BVd on OS beyond 36 months are uncertain. No definitive conclusion can be drawn regarding the effects of BVd treatment on HRQoL because of the high amount of missing data over time, the potential bias due to assessor knowledge of treatment assignment, and imprecision, making the direction of effects unclear.

Overall, the safety of BVd was consistent with the known safety profiles of the individual drugs for RRMM. However, the DREAMM-7 trial showed that treatment with BVd likely results in an increase in the proportion of patients who experience SAEs when compared with DVd. Notably, ocular AEs were more common and severe in the group receiving BVd, requiring more frequent treatment modifications. Despite these higher incidences of AEs in the BVd group, they were expected to be manageable with proper monitoring and supportive care in clinical practice. Further, unlike other therapies, belantamab does not exhibit significant myelosuppressive toxicities, making it a beneficial treatment option for older patients who are more vulnerable to the adverse effects of other treatments.

The results of the sponsor-submitted NMA suggested a favourable treatment effect of BVd on PFS compared to several other regimens, including IhKd, hKd, PVd, SVd, Kd, and Vd. Similarly, a favourable effect on OS was observed relative to hKd, PVd, SVd, and Vd. However, the magnitude of these effects remains uncertain.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of BVd, administered intravenously once every 3 weeks, at a dose of 2.5 mg/kg, in the treatment of RRMM in adults who have received at least 1 prior therapy.

Disease Background

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

MM is a progressive and incurable malignancy characterized by the clonal expansion of malignant plasma cells and the overproduction of M proteins.¹ Older individuals are more likely to develop MM, with a higher incidence in men compared to women; and in the US, it is more prevalent in African-American populations relative to people who are white and people of Asian ethnicity.^2,3 The incidence rate of MM increased in Canada from 2003 to 2019 by about 1.3% per year for females and 1.7% per year for males.⁴ In Canada, approximately 3,900 new MM cases were diagnosed in 2023 and there were an estimated 1,700 MM-related deaths.⁵ In 2024, around 4,100 new cases were projected, with 1,750 cases in females and 2,400 in males.⁶ The 5-year survival rate for patients with MM is approximately 50%,⁴ though recent therapeutic advancements have improved survival outcomes, despite the lack of a cure.^7,8

MM is most commonly diagnosed when patients present with symptoms such as bone pain, fatigue, anemia, kidney dysfunction, and recurrent infections.⁹ MM diagnosis requires the presence of 1 or more myeloma-defining events, along with either 10% or more clonal plasma cells in the bone marrow or a biopsy-proven plasmacytoma.³ Myeloma-defining events include end-organ damage (the CRAB criteria of hypercalcemia, renal insufficiency, anemia, and bone lesions) and 3 biomarkers: a clonal bone marrow plasma cell percentage of at least 60%, an FLC ratio of at least 100, and at least 1 focal lesion on an MRI.³ Prognosis is heavily influenced by staging, with the ISS and its revised version, the R-ISS, widely used to assess disease progression.^10,11 ISS evaluates albumin and B2M levels, where advanced stages correlate with low albumin and high B2M. The R-ISS further incorporates tumour burden and high-risk genetic abnormalities to refine prognostication.^10,11 Other factors, such as age, kidney function, and overall performance status, also impact prognosis.¹²

The build-up of resistance to different classes of therapies represents a significant challenge in the clinical management of MM because the disease typically becomes more resistant to treatment with each subsequent line of therapy.^13-15 In Canada, studies show that remission rates decrease with each line of therapy, and attrition rates increase.¹⁶ These findings underscore the progressive nature of the disease and the growing need for diverse therapeutic options that can effectively manage RRMM. Patients with RRMM often experience a persistent symptom burden, including fatigue, bone pain, and depression, which can severely affect their QoL.^17,18 The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of BVd, administered intravenously once every 3 weeks, at a dose of 2.5 mg/kg, in the treatment of RRMM in adults who have received at least 1 prior therapy.

Standards of Therapy

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

According to the clinician group and clinical experts consulted, the most important treatment goals for patients with MM are to prolong survival, delay disease progression, extend the DoR, improve QoL, and minimize side effects. Clinician groups and clinical experts consulted by CDA-AMC for this review indicated that initial therapy for patients with MM depends on whether patients are transplant-eligible or transplant-ineligible at diagnosis, which is aligned with the CDA-AMC Provisional Funding Algorithm.²⁸ According to the clinical experts and clinician groups, the majority of patients who are ineligible for transplant can be given daratumumab, lenalidomide, and dexamethasone as frontline therapy. A small number of patients will receive lenalidomide in combination with dexamethasone, and some other combination of drugs that does not contain daratumumab, such as lenalidomide in combination with bortezomib and dexamethasone, with lenalidomide continued until disease progression. According to the clinical experts and clinician groups, patients who are eligible for transplant can receive induction therapy with either cyclophosphamide in combination with bortezomib and dexamethasone or lenalidomide in combination with bortezomib and dexamethasone, and then lenalidomide maintenance until disease progression.

According to the clinical experts and clinician groups, the current approach to treatment of RRMM depends on several factors, including patient factors (i.e., age, comorbidity, prior toxicity), line of therapy, and prior therapies received. The clinical group and clinical experts indicated that treatment options for patients who have relapsed while on lenalidomide maintenance include such regimens as proteosome inhibitor (bortezomib, carfilzomib)–containing combinations such as SVd, IhKd, or Kd with or without cyclophosphamide. The clinical group and clinical experts emphasized that although triplet regimens are preferred for second-line and beyond treatment, patients who have progressed on lenalidomide alone tend to have a poorer prognosis. The general strategy for the second-line treatment and beyond includes the use of a drug that the patient has either not been exposed to previously or has demonstrated sensitivity to in the treatment regimen.²⁹

According to the Provisional Funding Algorithm for Multiple Myeloma developed by CDA-AMC,²⁸ patients with drug resistance cannot be treated again with the same drug; however, in later lines, previously used treatments are often reused when accessible due to a lack of novel options.³⁰ The clinical group indicated that in patients who have received anti-CD38s, IMiDs, and proteasome inhibitors, and were refractory to their last treatment, their options in fourth-line treatment and beyond include teclistamab, elranatamab, or ciltacabtagene autoleucel. The clinical experts consulted noted that the treatment landscape in Canada will undergo significant changes, with BCMA and CAR T-cell therapies being used in second-line treatment and beyond, and bispecific antibodies being used in fourth-line treatment and beyond.

Drug Under Review

BVd is indicated for the treatment of adults with RRMM who have received at least 1 prior line of therapy. The recommended starting dose schedule of BVd is 2.5 mg/kg administered intravenously once every 3 weeks for the first 8 cycles, and then continued as a single drug until disease progression or unacceptable toxicity.

Belantamab is a humanized, afucosylated, BCMA-targeted, antibody-drug conjugate that is fused to the microtubule inhibitor auristatin-F by a protease-resistant cysteine linker.^31-33 BCMA is an established therapeutic target for MM due to its highly selective expression on malignant plasma cells.^33-35 Belantamab binds to cell surface BCMA and is rapidly internalized into the tumour cell.³⁶ Upon internalization, the cytotoxic drug is released, disrupting the microtubule network and leading to cell cycle arrest and the apoptosis of tumour cells. The antibody enhances the recruitment and activation of immune effector cells, killing tumour cells by antibody-dependent cellular cytotoxicity and phagocytosis. Apoptosis induced by belantamab is accompanied by markers of immunogenic cell death, which may contribute to an adaptive immune response to tumour cells.^33-35,36

BVd underwent a standard review by Health Canada. The sponsor’s requested reimbursement criteria for BVd are aligned with the Health Canada indication.

Key characteristics of BVd are summarized in Table 3 with other treatments available for adult patients with MM who have received at least 1 prior therapy.

Table 3: Key Characteristics of Belantamab Mafodotin, Daratumumab, Isatuximab, Selinexor, Carfilzomib, Bortezomib, and Pomalidomide

ADC = antibody-drug conjugate; ADCC = antibody-dependent cell-mediated cytotoxicity; ADCP = antibody-dependent cellular phagocytosis; BCMA = B-cell maturation antigen; BVd = belantamab mafodotin plus bortezomib and dexamethasone; CD38 = Cluster of Differentiation 38; CDC = complement-dependent cytotoxicity; DRESS = Drug Rash with Eosinophilia and Systemic Symptoms; DVd = daratumumab plus bortezomib and dexamethasone; DVT = deep vein thrombosis; IhKd = isatuximab plus high-dose carfilzomib and dexamethasone; Kd = carfilzomib plus dexamethasone; MM = multiple myeloma; MMAF = monomethyl auristatin-F; PML = progressive multifocal leukoencephalopathy; PVd = pomalidomide plus bortezomib plus dexamethasone; q.w. = every week; RRMM = relapsed or refractory multiple myeloma; SC = subcutaneous; SCT = stem cell transplant; SJS = Stevens-Johnson syndrome; SVd = selinexor plus bortezomib plus dexamethasone; TEN = toxic epidermal necrolysis; TLS = tumour lysis syndrome; Vd = bortezomib plus dexamethasone; XPO1 = exportin 1.

^aHealth Canada–approved indication.

Sources: Product monographs for belantamab mafodotin,³⁷ Darzalex,³⁸ Sarclisa,³⁹ Velcade,⁴⁰ Xpovio,⁴¹ Pomalyst,⁴² and Kyprolis.⁴³

Perspectives of Patients, Clinicians, and Drug Programs

The full patient and clinician group submissions received are available in the consolidated patient and clinician group input document for this review on the project website.

Patient Group Input

This section was prepared by the review team based on the input provided by patient groups.

Patient input was submitted by 1 patient group: Myeloma Canada. Information was gathered from patients and caregivers through an online survey (N = 292). The most important symptoms related to myeloma that respondents want to control, in order of importance, include infections, mobility, renal problems, and pain. Forty-one percent of patient respondents indicated that they required the help of a caregiver to manage their disease or treatment-related symptoms. The input noted that patients’ disease symptoms impacted their ability to travel, work, exercise, and conduct volunteer activities. Respondents noted that the most significant financial implications due to myeloma treatment were the loss of income and/or pension funds due to absence from work, disability, or early retirement followed by the costs of travel, parking, drugs, and accommodation. Respondents felt that the interruption of life goals or accomplishments had the greatest impact on their QoL, followed by the loss of sexual desire, and anxiety or worry. Respondents reported a desire for a treatment that extends their life expectancy without disease progression (67% of respondents rated this “extremely important”) and improves their QoL (58% rated this “extremely important”).

No survey respondents had experience with BVd and 10 respondents indicated they had experience with belantamab (7 respondents with belantamab, pomalidomide, and dexamethasone, and 3 respondents either with belantamab as monotherapy or combined with dexamethasone). Among these respondents, the least bearable side effects reported included blurry vision, dry eyes, eye irritation, sensitivity to light, and infections. Eight of the respondents indicated that the overall side effects of belantamab treatment were “somewhat or mostly” manageable. When asked if the treatment was effective in controlling myeloma, 6 respondents reported “mostly or completely,” 3 respondents stated “somewhat,” and 1 respondent stated “slightly.” The input emphasized the importance of patient preference in weighing the potential costs and benefits of any new treatment.

Clinician Input

Input From Clinical Experts Consulted for This Review

All CDA-AMC 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, providing guidance on the potential place in therapy). The following input was provided by 3 clinical specialists with expertise in the diagnosis and management of MM.

Unmet Needs

Despite significant advances in the treatment of MM, substantial unmet needs remain in its management. The clinical experts consulted by CDA-AMC highlighted several challenges associated with the currently available therapies for MM in Canada. First, while various classes of drugs are available to manage MM at different stages, no single treatment can offer a definitive cure for the disease. Second, a critical treatment gap exists for patients who become refractory to lenalidomide or anti-CD38 therapies. This is especially problematic for older adults who have previously received daratumumab in combination with lenalidomide plus dexamethasone because limited therapeutic options are available upon relapse. Third, treatment regimens such as Kd or SVd demonstrate suboptimal outcomes, and many patients may not be candidates for more aggressive therapies like CAR T-cell therapy or bispecific antibodies. Additionally, there is a need for therapies with improved toxicity profiles. Because patients with MM often require treatment regimens with manageable side effects to preserve their QoL, there is an increasing demand for therapies that involve fewer clinic visits and less intensive monitoring.

Place in Therapy

The clinical experts consulted by CDA-AMC indicated that belantamab would primarily be used in second-line treatment and beyond for MM, specifically for patients who have relapsed or are refractory to prior therapies. The clinical experts indicated that as a BCMA-targeting therapy, belantamab offers a novel mechanism of action that distinguishes it from other BCMA-targeted therapies. The clinical experts noted that for patients who are refractory to lenalidomide and proteasome inhibitors, belantamab could be considered in the early relapse phase (after 1 to 3 prior lines of therapy). The clinical experts noted that for patients who have not been exposed to anti-CD38 therapy, the current standard of care for those who can tolerate it includes isatuximab plus carfilzomib and dexamethasone, isatuximab plus pomalidomide and dexamethasone, or DVd. The clinical experts indicated that BVd could potentially be used as an alternative to SVd, Kd, and pomalidomide plus dexamethasone in older patients. However, in younger patients, if the ciltacabtagene autoleucel eligibility criteria require no prior BCMA exposure, the use of BVd will likely be minimal in these settings.

Patient Population

The clinical experts indicated that older patients, particularly those who have previously received lenalidomide plus bortezomib and dexamethasone, or daratumumab in combination with lenalidomide and dexamethasone, and are not refractory to bortezomib, are ideal candidates for belantamab. They noted that these patients are often undertreated with the current standard-of-care options, making belantamab a promising alternative. The clinical experts indicated that in younger adults, the availability of ciltacabtagene autoleucel in second-line and beyond treatment may limit the use of belantamab, especially if they have had prior exposure to BCMA therapies. However, the clinical experts noted that belantamab may still be considered for these patients in later relapse settings. The clinical experts emphasized that, aside from ocular toxicities, the toxicity profile of belantamab is relatively favourable, particularly in older adult patients. Unlike other therapies, belantamab does not exhibit significant myelosuppressive toxicities, making it a good option for older patients who are more vulnerable to the adverse effects of other treatments. However, the clinical experts indicated that ocular toxicities remain a concern and patients with pre-existing ocular conditions should be carefully assessed for eligibility and closely monitored for ocular toxicity. The clinical experts noted that belantamab has demonstrated profound and sustained responses, with clinical trials suggesting that dose reductions and extended intervals between infusions preserve efficacy while providing good tolerability, including in relation to ocular toxicity.

Assessing the Response Treatment

The clinical experts noted that assessing response to treatment for MM is generally straightforward, with routine blood tests being the primary method of monitoring. These include complete blood count, chemistry panels, FLC assays, serum protein electrophoresis, and immunofixation electrophoresis, which are typically performed every 4 weeks or before each new treatment cycle. Bone imaging is used less frequently but can be helpful for assessing response. The clinical experts indicated that improvement in MM is closely associated with an enhancement in QoL. However, ocular toxicity with belantamab therapy can negatively impact QoL and requires more intensive monitoring than is standard in care for MM. Specifically, regular ophthalmologic assessments and slit lamp examinations are necessary to guide dose adjustments and prevent more severe ocular toxicity because they are not typically part of standard clinical practice in MM.

Discontinuing Treatment

The clinical experts indicated that treatment with BVd would be discontinued for 2 primary reasons: disease progression and intolerable toxicity. Disease progression, assessed though routine clinical testing and according to IMWG criteria, is the most common reason for discontinuation. The second major reason is intolerance to toxicities, with ocular toxicity being the most significant. The clinical experts indicated that ocular toxicity may require extending dosing intervals to every 8 weeks to 12 weeks, and in severe or intolerable cases, treatment may be discontinued. The clinical experts also noted that other toxicities, such as recurrent infections (especially if hospitalization is required), cytopenia, secondary primary malignancies, or intractable diarrhea, could also lead to discontinuation. The clinical experts emphasized the importance of establishing clear criteria for dose reductions or delays to manage toxicities effectively and reduce unnecessary treatment discontinuations.

Prescribing Considerations

The clinical experts emphasized that while community sites can manage myeloma treatment, the ocular toxicity associated with belantamab therapy requires specialized support from eye specialists, such as ophthalmologists or optometrists, depending on availability. Therefore, the treatment of MM should be managed by trained hematologist-oncologists, with eye specialists involved in monitoring and managing ocular toxicity. The clinical experts also noted that protective contact lenses are being tested as a potential way to mitigate the severity of ocular toxicity. Additionally, they mentioned that cytopenias and other adverse effects from treatments like bortezomib are typically well managed in both tertiary and community care settings.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups.

Two clinician groups consisting of a total of 27 clinicians provided input for this review, including the OH (CCO) Hematology Cancer Drug Advisory Committee and the CMRG. Both submissions noted that myeloma remains incurable despite the introduction of new drugs over the last 2 decades. Patients eventually become refractory to all available funded antimyeloma drugs. The input stated that symptom burden for patients with MM is high, with patients experiencing bone pain and destruction, anemia and other cytopenias, renal damage, hypercalcemia, and a high risk of infection. The CMRG stated that despite the clear benefits of lenalidomide as part of first-line therapy, progression on this potent drug even as single-drug maintenance leads to shorter PFS outcomes with nearly all traditional and reimbursed second-line regimens (including those containing an anti-CD38 monoclonal antibody) compared to the results without such exposure. Consequently, the input noted that drug exposure, rather than lines of therapy, more accurately defines the need for access to innovative treatments to forestall the development of refractory myeloma. The CMRG also noted that with the movement of combinations of 3 major drug classes (i.e., an IMiD, a proteasome inhibitor, and an anti-CD38 monoclonal antibody) to the first-line and second-line treatment setting, exposure and resistance to multiple drug classes now occurs much earlier in the disease course than in the past. Therefore, the highest unmet need in myeloma continues to be effective treatment for patients who have progressed despite exposure to effective drugs.

OH (CCO) noted that the regimen would be another second-line treatment option for patients who are sensitive to bortezomib. The input noted that the treatment under review would be most suitable for patients who are unlikely to receive CAR T-cell therapy as BVd may preclude the future use of BCMA-targeted CAR T-cell therapy. Standard myeloma response outcomes used in clinical practice would be appropriate to determine if a patient is responding to the treatment under review, based on the M protein markers in the serum and/or urine, a bone marrow biopsy and, in some instances, imaging studies. Clinically meaningful responses usually correlate with at least a partial remission as defined by IMWG consensus criteria. These include improvement in symptoms (the cessation of bone destruction with less pain, fewer fractures, and less need for radiotherapy), improvement in energy, and a better ability to perform activities of daily living. Responses are generally assessed every 1 to 3 months depending on clinical stability and the regimen used for therapy. Factors to consider when deciding to discontinue the treatment under review include significant toxicities, particularly ocular AEs, as well as disease progression. The input stated that the appropriate setting for BVd treatment is an outpatient setting and there is also a need for ophthalmological assessments.

Drug Program Input

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

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

Anti-CD38 = anti–Cluster of Differentiation 38; BCMA = B-cell maturation antigen; BVd = belantamab mafodotin plus bortezomib and dexamethasone; BPd = belantamab mafodotin plus pomalidomide and dexamethasone; CDA-AMC = Canada’s Drug Agency; cilta-cel = ciltacabtagene autoleucel; DVd = daratumumab plus bortezomib and dexamethasone; IhKd = isatuximab plus high-dose carfilzomib and dexamethasone; Kd = carfilzomib plus dexamethasone; PAG = Provincial Advisory Group; pERC = pan-Canadian Oncology Drug Review Expert Review Committee; PFS = progression-free survival; PVd = pomalidomide plus bortezomib plus dexamethasone; SVd = selinexor plus bortezomib plus dexamethasone.

Clinical Evidence

The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of BVd, administered intravenously once every 3 weeks, at a dose of 2.5 mg/kg, in the treatment of RRMM in adults who have received at least 1 prior therapy. The focus has been placed on comparing BVd to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of BVd is presented in 2 sections with critical appraisal of the evidence by CDA-AMC included at the end of each section. The first section, the systematic review, includes an RCT that was selected according to the sponsor’s systematic review protocol. The CDA-AMC assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes indirect evidence from the sponsor. No long-term extension studies or studies considered to address important gaps in the evidence were submitted by the sponsor.

Included Studies

Clinical evidence from the following is included in the review and appraised in this document:

• 1 RCT identified in the systematic review

• 1 ITC.

Systematic Review

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Studies

Characteristics of the included studies are summarized in Table 5.

Table 5: Details of Studies Included in the Systematic Review

AE = adverse event; anti-BCMA = anti–B-cell maturation antigen; anti-CD38 = anti–Cluster of Differentiation 38; ASCT = autologous stem cell transplant; AUC = area under the curve; BVd = belantamab mafodotin plus bortezomib and dexamethasone; CBR = clinical benefit rate; C_max = peak concentration; CRR = complete response rate; DoR = duration of response; DVd = daratumumab plus bortezomib and dexamethasone; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EORTC IL52 = European Organisation for Research and Treatment of Cancer Item Library 52; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EORTC QLQ-MY20 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Multiple Myeloma 20; FACT-GP5 = Functional Assessment of Cancer Therapy–Item GP5; FLC = free light chain; HRQoL = health-related quality of life; IA1 = interim analysis 1; IA2 = interim analysis 2; IMWG = International Myeloma Working Group; MM = multiple myeloma; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; OSDI = Ocular Surface Disease Index; PFS = progression-free survival; PFS2 = progression-free survival-2; PGI-C = Patient Global Impression–Change; PGI-S = Patient Global Impression–Severity of Illness; PRO-CTCAE = Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events; SAE = serious adverse event; TTBR = time to best response; TTP = time to progression; TTR = time to response; VGPR = very good partial response; vs. = versus.

^aInduction therapy followed by ASCT and maintenance constitutes 1 line of therapy.

^bPatients who have undergone syngeneic transplant were allowed, only if there was no history of graft-vs.-host disease.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

The DREAMM-7 study²⁶ is an ongoing, phase III, randomized, open-label, multicentre trial that aims to assess the efficacy and safety of belantamab administered intravenously at the dose of 2.5 mg/kg on day 1 of every 21-day cycle, in combination with bortezomib and dexamethasone (together known as BVd), compared with DVd in adult patients with MM. The trial enrolled patients who had received at least 1 prior line of MM therapy and had documented disease progression during or after their most recent treatment. Patients were enrolled in the trial across 142 centres in 20 countries, including 5 sites in Canada.

The DREAMM-7 study included a screening period, a treatment period, and a follow-up period (Figure 1). During the screening period, patients were evaluated for study eligibility according to the protocol. Following screening, patients were centrally randomized in a 1:1 ratio through an Interactive Response Technology system to receive either BVd (N = 243) or DVd (N = 251). Randomization was stratified by the number of prior lines of therapy (1 versus 2 or 3 versus more than 4), prior bortezomib use (yes versus no), and R-ISS (stage I versus stage II or stage III). No more than 50% of patients with 2 or more prior lines of therapy were enrolled, and crossover treatment was not permitted.

During the treatment period, safety and disease assessments were performed regularly for each treatment group. Treatment continued in both groups until disease progression, death, unacceptable toxicity, a decision at the investigator’s discretion, withdrawal of consent, or the end of the study, whichever occurred first. For patients who discontinued study treatment for reasons other than disease progression or death, disease evaluations for PFS were performed every 3 weeks until documented disease progression, death, the start of a new antimyeloma treatment, the withdrawal of consent, a loss to follow-up, or the end of the study, whichever occurred first. In the event of disease progression, patients were followed to ascertain the receipt of subsequent antimyeloma therapy, PFS-2, and survival status every 12 weeks until the withdrawal of consent, a loss to follow-up, death, or the end of the study. A patient was considered to have completed the study if they were followed until death or the end of the study. The end of treatment visit occurred within 30 days of the last cycle or before the initiation of new antimyeloma therapy, whichever occurred first.

The primary objective of the DREAMM-7 trial was to compare the efficacy of BVd to DVd based on PFS. The key secondary objective was to compare the efficacy of BVd to DVd based on OS, DoR, and the MRD negativity rate. The primary and secondary end points of the DREAMM-7 trial were analyzed using data from the cut-off date of October 2, 2023, and the database lock date of November 6, 2023, for IA1, and from the cut-off date of October 7, 2024, for IA2 or prespecified primary PFS analysis.

Figure 1: Study Design of the DREAMM-7 Trial

OS = overall survival; PD = progressive disease; PFS = progression-free survival; Q1W = every week; Q3W = every 3 weeks; Q4W = every 4 weeks; R-ISS = Revised International Staging System; SC = subcutaneous.

^a Stratification: Prior lines of therapy (1 versus 2 or 3 versus ≥ 4), R-ISS stage I versus R-ISS stage II or stage III, and prior bortezomib use (yes versus no).

^b Reduce starting dose of dexamethasone to 10 mg for patients aged older than 75 years, for those with a body mass index of less than 18.5 kg/m², for those who had previous unacceptable side effects associated with glucocorticoid therapy, or for those who were unable to tolerate the starting dose.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Populations

Inclusion and Exclusion Criteria

Patients were eligible for enrolment in the DREAMM-7 trial if they were aged 18 years or older with a confirmed diagnosis of MM as defined by IMWG criteria.⁴⁴ Patients were required to have an ECOG PS score of 0 to 2, have received at least 1 prior line of MM therapy with documented disease progression during or after their most recent therapy, and have at least 1 measurable disease parameter. Key exclusion criteria included intolerance to daratumumab and bortezomib, and refractoriness to either daratumumab or any anti-CD38 therapy or bortezomib (on a twice-weekly regimen). Patients were excluded from the trial if they had received prior treatment with an anti-BCMA therapy or a monoclonal antibody within 30 days of receiving the first dose of study drugs.

Interventions

Intervention Group (BVd)

Belantamab was administered intravenously at the dose of 2.5 mg/kg on day 1 of every 21-day cycle for the first 8 cycles and then continued as a single drug until confirmed disease progression, unacceptable toxicity, death, the withdrawal of consent, or study end, whichever occurred first (Figure 2). Belantamab was administered after dilution as an IV infusion over approximately 30 minutes. Prophylaxis to mitigate ocular events was instituted for all patients.

Bortezomib was administered subcutaneously at a dose of 1.3 mg/m² on day 1, day 4, day 8, and day 11 of each 21-day cycle for a total of 8 cycles. The administration of bortezomib was scheduled approximately 1 hour following the completion of the belantamab infusion.

Dexamethasone 20 mg (either orally or intravenously) was administered on the day of and the day after bortezomib treatment. On days when bortezomib and dexamethasone administration coincided with the administration of belantamab, dexamethasone was administered orally or intravenously before the infusion of belantamab. The starting dose of dexamethasone was reduced to 10 mg for patients aged older than 75 years, for those with a body mass index of less than 18.5 kg/m², for those who had previously experienced unacceptable side effects from glucocorticoid therapy, or for those unable to tolerate the standard starting dose.

Figure 2: Dosing Schedule for the Intervention Group (BVd) in the DREAMM-7 Trial

BVd = belantamab mafodotin in combination with bortezomib and dexamethasone.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Dose delays and reductions were permitted throughout the study.⁴⁵ Patients were assessed by a qualified eye care specialist at screening, and then every 3 weeks before dosing up to the sixth dose of belantamab. A cooling eye mask was applied during the first few hours after belantamab administration to help reduce ocular side effects. Corticosteroid eye drops were not required but could be used if clinically indicated at the discretion of an eye care specialist. Preservative-free artificial tears were administered at least 4 times to 8 times daily in each eye beginning on cycle 1, day 1, until the end of treatment. In the event of ocular symptoms (e.g., dry eyes), the use of artificial tears could be increased up to every 2 hours, as needed.

Comparison Group (DVd)

Daratumumab was administered intravenously at a dose of 16 mg/kg according to the approved label schedule as follows: weekly for cycle 1 through cycle 3 (week 1 to week 9) in 21-day cycles for a total of 9 doses, on day 1 of cycle 4 through cycle 8 (week 10 to week 24) in 21-day cycles for total of 5 doses, and then every 4 weeks starting from cycle 9 (week 25) onward in 28-day cycles. For the first dose of daratumumab (administered only at week 1), the single infusion could be split over 2 days (Figure 3).

The dosing schedule for bortezomib and dexamethasone was identical to that of the intervention group. Belantamab and daratumumab were administered intravenously to patients at the site while bortezomib was administered subcutaneously to patients at the site.

Figure 3: Dosing Schedule for Comparator Group (DVd) in the DREAMM-7 Trial

DVd = daratumumab plus bortezomib and dexamethasone.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Dose Modification, Delay, and Study Treatment Discontinuation

Dose modifications, delays, and reductions for belantamab were permitted throughout the DREAMM-7 trial as follows: the dose of belantamab could be reduced once from the starting dose (2.5 mg/kg intravenously every 21 days) to 1.9 mg/kg intravenously every 21 days (Table 6). Patients who were unable to tolerate a dose of 1.9 mg/kg were permanently discontinued from belantamab treatment due to unacceptable toxicity.

Table 6: Dose Modification of Belantamab Mafodotin for Ocular Adverse Reactions

BCVA = best corrected visual acuity.

^aNonocular adverse reactions were graded according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events.

^bA corneal defect may lead to corneal ulcers. These should be managed promptly and as clinically indicated by an eye care professional.

Sources: Product monograph for belantamab mafodotin.³⁷ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

No dose modifications for daratumumab were allowed in the DREAMM-7 trial. Bortezomib dose adjustments were based on the highest grade of toxicity attributed to bortezomib. Bortezomib therapy should be withheld upon the onset of any grade 3 nonhematological or grade 4 hematological toxicities, excluding neuropathy. Once the toxicity symptoms have resolved, bortezomib therapy could be reinitiated at a reduced dose (1.0 mg/m² to 0.7 mg/m²). For patients with contraindications to the starting dose regimen of dexamethasone (20 mg or 10 mg) in combination with bortezomib, or who were intolerant to this regimen, the dose of dexamethasone could be reduced to 12 mg for patients aged 75 years or younger, or 10 mg for those aged older than 70 years with a body mass index less than or equal to 18.5 kg/m². If the reduced dose was not tolerated, the dose of dexamethasone could be further reduced to 8 mg and 4 mg, respectively. If these doses are also not tolerated, dexamethasone should be discontinued.

The study drug must be permanently discontinued in the event of disease progression, unacceptable toxicity, the patient meeting any of the protocol-defined safety stopping criteria, or pregnancy. Patients who had permanently discontinued study treatment were not allowed to resume treatment. All patients who had permanently discontinued study treatment were to undergo safety assessments at the time of discontinuation, during the end of the treatment visit, and throughout the follow-up period.

Concomitant Medications

Concomitant therapy with bisphosphonates was allowed and recommended. Patients were to receive full supportive care during the study, including transfusions of blood products, growth factors, and treatment with antibiotics, antiemetics, antidiarrheals, and analgesics, as appropriate. The following concomitant medications were prohibited during the DREAMM-7 trial: live vaccines or live, attenuated vaccines, plasmapheresis, other antimyeloma therapy not specified in the protocol, inhibitors of P-glycoprotein, and organic anion transporting polypeptide. Additionally, for patients receiving bortezomib, the use of organic anion transporting polypeptide inhibitors, anti-HIV and anti–hepatitis C virus drugs, antibiotics, and antifungals was prohibited.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 7, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review, according to the clinical experts consulted for this review and input from patient and clinician groups and public drug plans. Using the same considerations, CDA-AMC selected the end points that were considered to be most relevant to inform expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. The end points considered most important to the expert committee deliberations (OS, PFS, HRQoL, ocular events, and SAEs) were assessed using the GRADE tool. Other efficacy end points (ORR, CRR, DoR, and MRD negativity) are summarized in the report as supportive information.

Response evaluation was performed according to the IMWG Uniform Response Criteria for Multiple Myeloma⁴⁶ (Table 8), as determined by a blinded IRC.

Table 7: Outcomes Summarized From the DREAMM-7 Trial Included in the Systematic Review

CR = complete response; CRR = complete response rate; DoR = duration of response; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HRQoL = health-related quality of life; IA1 = interim analysis 1; IA2 = interim analysis 2; IMWG = International Myeloma Working Group; MR = minimal response; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; PD = progressive disease; PFS = progression-free survival; PR = partial response; sCR = stringent complete response; VGPR = very good partial response.

Note: All categories of disease response used in the calculation of study end points (sCR, CR, VGPR, MR, PR, stable disease, and PD) were determined by an independent review committee using IMWG 2016 criteria.

^aAn outcome that is commonly used in clinical practice to assess patients.

^bAn outcome that is considered important to patients according to patient group input and/or clinician input.

^cStatistical testing for these end points was adjusted for multiple comparisons.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 8: IMWG Criteria for Response Assessment

CR = complete response; FLC = free light chain; IMWG = International Myeloma Working Group; MR = minimal response; PD = progressive disease; PR = partial response; sCR = stringent complete response; SD = stable disease; SPD = sum of the products of diameters; VGPR = very good partial response.

Sources: Kumar et al. (2016).⁴⁷ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Progression-Free Survival

The primary outcome for the DREAMM-7 trial was PFS, which was defined as the time from the date of randomization until the earliest date of documented disease progression determined by an IRC, according to IMWG criteria,²⁵ or death due to any cause. Results of PFS are available for both IA1 and IA2. Disease assessments were conducted every 3 weeks after treatment discontinuation for reasons other than disease progression.

In the DREAMM-7 trial, the reasons for censoring for PFS included no disease progression or being alive at the data cut-off date for the first or second interim analyses (IA1 or IA2), the absence of an adequate baseline or postbaseline assessment, no disease progression or being alive with an adequate postbaseline assessment, the initiation of new antimyeloma treatment before documented disease progression with or without an adequate postbaseline assessment, and death or disease progression after missing 2 or more scheduled assessments.

Overall Survival

OS was a key secondary end point in the DREAMM-7 study and was defined as the time from the date of randomization until the date of death due to any cause. Patients who were alive at the data cut-off date were censored at the date of last contact. Assessments for survival were conducted every 3 weeks after treatment discontinuation for reasons other than disease progression, and every 12 weeks after disease progression. The results of OS are available for both IA1 and IA2.

Duration of Response

DoR was a key secondary end point in the DREAMM-7 trial and was defined as the time from the first documented evidence of PR or better until PD or death due to any cause, based on an IRC assessment per IMWG criteria. In the DREAMM-7 trial, responders without disease progression were censored at the censoring time point for time to progression. The reasons for censoring for DoR included being alive with or without an adequate baseline assessment, being alive and having no disease progression with postbaseline assessment, the absence of postbaseline assessment before the initiation of new antimyeloma treatment, the initiation of new antimyeloma treatment before documented disease progression with an adequate postbaseline assessment, and disease progression with or without death after missing 2 or more scheduled assessments.

MRD Negativity

MRD was a key secondary end point in the DREAMM-7 trial and was defined as the percentage of patients who achieved MRD-negative status (as assessed by next-generation sequencing at a 10^-5 threshold) at least once during the time of confirmed CR or better response (i.e., CR or sCR), based on an IRC assessment per IMWG criteria. The US FDA considers MRD a biomarker that is a reliable quantification of tumour burden, independent of the assay used.¹⁹ In MM, the FDA recommends that MRD should be assessed only in patients who are in CR, which aligns with the criteria used in the DREAMM-7 trial. After the initial MRD sample collection during the screening period, subsequent MRD testing was to be conducted when a patient first achieved a confirmed response of very good partial response (VGPR), and then every 6 months from the time of the last sample collection until disease progression or suspected CR.

Complete Response Rate

CRR was a secondary end point in the DREAMM-7 trial and was defined as the percentage of patients with a confirmed CR or better response (i.e., CR or sCR), based on an IRC assessment per IMWG criteria. While both CR and sCR require the absence of detectable M protein in serum and urine, as well as fewer than 5% of plasma cells in the bone marrow, sCR also requires a normal FLC ratio and the absence of clonal cells detectable by immunohistochemistry, indicating a deeper level of response.

Overall Response Rate

ORR was a secondary end point in the DREAMM-7 trial and was defined as the percentage of patients with a confirmed PR or better (i.e., VGPR, CR, or sCR), based on an IRC assessment per IMWG criteria. ORR is a direct measure of antitumour activity and is a well-established end point in oncology clinical trials,^48,49 having been used as a secondary end point in pivotal clinical trials of other approved systemic therapies in RRMM.^50-52

HRQoL

HRQoL was a secondary end point in the DREAMM-7 trial and was measured by the change in baseline in the EORTC QLQ-C30 score.

The EORTC QLQ-C30 module is a 30-item questionnaire containing both single- and multi-item measures to assess HRQoL in patients living with cancer. These 30 items can be categorized into the following: 1 GHS or QoL scale, 5 functional scales, 3 symptom scales, and 6 single-item scales. Each scale is scored from 0 to 100, with a higher score representing more of the concept (e.g., more functioning, more symptoms). Each of the multi-item scales includes a different set of items. No item occurs in more than 1 scale. Here is a breakdown of the scales:

• 1 GHS or QoL scale (2 items)

• 5 functional scales — physical functioning (5 items), role functioning (2 items), emotional functioning (4 items), cognitive functioning (2 items), and social functioning (2 items)

• 3 symptom scales — fatigue (3 items), nausea and vomiting (2 items), and pain (2 items)

• 6 single-item scales relating to dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties.

The validity, reliability, and minimal important difference of the EORTC QLQ-C30 are summarized in Table 9.

Harms Outcomes

Safety and tolerability were assessed at each visit based on the incidence of AEs, SAEs, deaths, AEs leading to discontinuation, AEs leading to dose reduction or interruption, and AEs of special interest. AEs were coded using the standard Medical Dictionary for Regulatory Activities and graded by the investigator according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events version 5 (CTCAE v5.0), or the KVA scale, as appropriate.⁵³ All SAEs were collected from the initiation of the study intervention until at least 70 days after the discontinuation of all study interventions. Any SAEs considered related to study participation (e.g., study intervention, change in existing therapy) were recorded from the time the patient consented to participate in the study through any follow-up period.

The AEs of special interest for belantamab included ocular events, corneal events, thrombocytopenia, and infusion-related reactions. Corneal events associated with belantamab were graded using the KVA scale while other treatment-related ocular AEs were assessed according to the CTCAE v5.0 criteria for eye disorders. The severity of all other AEs of special interest was graded using CTCAE v5.0.⁵³

Table 9: Summary of Outcome Measures and Their Measurement Properties

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HRQoL = health-related quality of life; HSCT = hematopoietic stem cell transplant; MID = minimal important difference; MM = multiple myeloma; QoL = quality of life; SRM = standardized response mean; vs. = versus.

^aThe EBMT criteria for response were used to determine the patients’ disease phase.

^bTo assess the magnitude of the difference in scores between patients who improved, deteriorated, and remained stable, SRMs were calculated and compared against the Cohen rule of thumb for interpreting the magnitude of mean differences in HRQoL scores, with 0.20 representing a small change, 0.50 a moderate change, and more than 0.80 a large change.

^cThe MIDs were estimated using an anchor-based approach; they were anchored to a structured QoL interview (response options were improved, deteriorated, or unchanged).

Statistical Analysis

A summary of the statistical analyses performed for all efficacy end points is presented in Table 10.

Sample Size and Power Calculation

The sample size was driven by the analysis of the primary end point, PFS. Based on data from the CASTOR study,⁶⁴ it was estimated that approximately 259 events (disease progression or death) were needed to ensure 90% power to detect a significant difference between the BVd group and the DVd group in PFS, at a 1-sided significance level of 2.5%. It was estimated that the targeted 259 PFS events would be observed approximately 33 months from the time when the first patient was randomized under the alternative hypothesis, assuming an annual dropout rate of 5%.

The DREAMM-7 study was designed to include 3 interim analyses followed by a final analysis. The first interim analysis (IA1) was planned at the time when approximately 250 of the 280 targeted PFS events (an approximately 89% information fraction) had been observed. If PFS demonstrated statistical significance at IA1, the rationale for IA2 would be driven by the requirements for OS (PFS would not be retested). The timing of subsequent analyses would be determined based on the OS information fraction.

Using available data from the CASTOR study, the median OS in the DVd group was expected to be around 49 months. With 478 patients, the power to test the null hypothesis for OS (HR = 1 versus the specific alternative hypothesis HR = 0.85) was approximately 16%. It was estimated that approximately 137 deaths would be observed at the time of the primary PFS analysis. To ensure 80% power to test the null hypothesis for OS, a total of 355 deaths needed to be observed. The information fraction may shift dependent on the actual timing of analyses and the observed OS events at that time and the boundaries may be adjusted accordingly.

Table 10: Statistical Analysis of Efficacy End Points in the DREAMM-7 Trial

CMH = Cochran-Mantel-Haenszel; CRR = complete response rate; DoR = duration of response; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; KM = Kaplan-Meier; mITT = modified intention to treat; MMRM = mixed model of repeated measures; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; RMDoR = restricted mean duration of response; VGPR = very good partial response.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Multiple Comparisons and Multiplicity

At the time of IA1, using the October 2, 2023, data cut-off, the familywise type I error rate for testing multiple end points was strongly controlled with a step-down or hierarchical testing procedure. The hierarchical multiple-testing strategy is presented in Figure 4.

The evaluation of the primary and key secondary end points was structured in terms of 2 families of hypotheses as follows.^26,65 The first family was based on the primary end point of PFS at IA1 and IA2. IA1 was performed with a multiplicity-adjusted boundary for a significance of P of less than 0.018. The Lan-DeMets approach,⁶⁶ which approximates the O’Brien-Fleming spending, was used to maintain an overall 1-sided, 2.5% type I error when testing PFS across IA1 and the primary PFS analysis or IA2.^26,65

The second family was based on the 3 key secondary end points: OS, DoR (assessed using RMDoR), and MRD negativity. OS was planned to be evaluated through 4 planned analyses: IA1, the primary PFS analysis or IA2, a third interim analysis (interim analysis 3 or IA3), and the final OS analysis. IA3 is planned to occur after approximately 266 OS events (an approximately 75% OS information fraction), and the final OS analysis will be conducted after approximately 355 OS events (a 100% OS information fraction). Given that the testing of the first family was successful (P ≤ 0.018 for PFS), the full significance level (alpha = 0.025) was propagated to the second family of hypotheses. For the second family, a weighted Bonferroni procedure was applied across OS and DoR, where alpha was split between the end points, with a larger proportion initially assigned to OS. At the time of IA1, for the second family, DoR (assessed using RMDoR) was significant (P value ≤ 0.005) but OS was not significant (P value > 0.00037); therefore, OS was analyzed descriptively. Additionally, MRD negativity testing was contingent upon the successful rejection of the null hypothesis for OS, in accordance with a hierarchical testing procedure. However, given that OS was not successful, MRD negativity was analyzed descriptively.

The analysis of key secondary end points DoR and MRD negativity was based on data available at the time of IA1 in accordance with the prespecified statistical analysis plan. At the time of IA2, PFS was analyzed descriptively without formal testing, based on the data available at the data cut-off. Regardless of the timing of PFS statistical significance, formal testing of DoR and MRD negativity should be based on data at the time of IA1. At the time of IA2, the key secondary end points DoR and MRD negativity were analyzed descriptively, without formal testing.

Figure 4: Multiple-Testing Strategy in the DREAMM-7 Trial

DoR = duration of response; H₁ = hypothesis 1; H₂ = hypothesis 2; H₃ = hypothesis 3; H₄ = hypothesis 4; IA1 = interim analysis 1; IA2 = interim analysis 2; IA3 = interim analysis 3; MRD = minimal residual disease; OS = overall survival; PFS = progression-free survival.

Note: Hi denotes the 1-sided null hypothesis for the primary and key secondary end points, where i equals 1,2,3,4 denotes the index indicating PFS, OS, DoR, and MRD negativity rate, respectively. Upon the successful rejection of the hypothesis and regardless of the timing of the rejection, the full alpha allocated to testing the hypothesis can be propagated. Arrows indicate the direction and proportion of alpha reallocation. H₁ was to be tested at the 1-sided 2.5% significance level. All other hypotheses had an initial alpha of 0% assigned. The number of rectangular boxes indicates the number of planned analyses with alpha allocation for a given hypothesis, with text indicating the corresponding end point and time point of data extraction to be tested. Alpha was to be adjusted to account for multiple testing of an end point across time points using the Lan-DeMets approach that approximates the O'Brien-Fleming spending function. The efficacy boundaries were to be adjusted based on the observed number of events at the time of analysis.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Statistical Analysis

PFS and OS

The analysis of time-to-event outcomes PFS and OS was conducted using the ITT population. A stratified Cox proportional hazards model with the Efron method of tie handling and treatment group as the sole explanatory variable was used to assess the magnitude of the treatment difference (HR) in PFS and OS between the treatment groups. The treatment difference was compared using the log-rank test at a 1-sided alpha level of 0.025, stratified by randomization factors. The nonparametric KM method was employed to estimate survival curves for PFS, with KM curves presented by treatment group. The PFS rate at 6 months, 12 months, 18 months, and 24 months with a corresponding 95% CI was also estimated from the KM analysis, using the Brookmeyer-Crowley method.⁶⁷ The type of events (progressions and deaths) and censoring reasons were summarized.

The proportional hazards assumption was assessed using the following methods: a KM plot by treatment group, a plot of log(time) against log(-log[survival]) by treatment group, a plot of Schoenfeld residuals for treatment, and an evaluation of time dependency of treatment effect by adding an interaction term of treatment and time in the Cox model.

The following sensitivity analyses for PFS were performed at the time of the primary analysis (IA1):

• the non–proportional hazards effect on PFS, using —

  • the restricted mean survival time (RMST) method⁶⁸ to account for the possible non–proportional hazards effect

  • RMST as the expected survival time restricted to a specific time horizon common punctuation time (t*)

  • the cut-off t* for determining the RMST was the smallest value among the largest observed times across study interventions

• the PFS primary estimand based on investigator-assessed response, including only the primary estimand of PFS (i.e., the handling of intercurrent events premised on primary events and censoring rules) based on the investigator-assessed response

• the PFS primary estimand based on stratification data from the clinical database; premised on the presence of mis-stratification for stratification factors used for randomization, a supplementary analysis was performed using the stratification data based on the clinical database

• the PFS primary estimand based on a modified ITT analysis set using an IRC-assessed response

• the PFS primary estimand considering local efficacy laboratories.

The following additional analyses of supplementary PFS estimands were conducted:

• a primary analysis of supplementary estimand 1 (an IRC-assessed response and alternative censoring rules 1 [when progression was documented between scheduled visits, the event date was considered as the minimum date of the next scheduled visit or the date of death])

• a primary analysis of supplementary estimand 2 (an IRC-assessed response and alternative censoring rules 2 [the date of starting new antimyeloma therapy was considered as an event])

• a primary analysis of supplementary estimand 3 (an IRC-assessed response and alternative censoring rules 3 [when death or progression occurred after an extended loss to follow-up, this was considered as an event at the date of death or progression])

• a primary analysis of supplementary estimand 4 (an IRC-assessed response and alternative censoring rules 4 [treatment discontinuation due to clinical progression before progressive disease or death was considered as an event at the date of treatment discontinuation])

• a primary analysis of the COVID-19 supplementary estimand (an IRC-assessed response and COVID-19 censoring rules).

The following sensitivity analyses for OS were performed:

• the non–proportional hazards effect on OS — the RMST method was conducted to account for the possible non–proportional hazards effect

• adjusting for the poststudy therapy on OS to obtain a less biased estimate of treatment effect, an inverse probability of censoring weighting method was performed for OS to adjust for poststudy therapy.⁶⁹

Subgroup Analyses of PFS

Subgroup analyses were performed using the same methods as for the primary PFS analysis but only including patients within the relevant subgroup category. The following subgroup analyses were performed to compare the primary estimand of PFS between treatment groups, identified as important by clinical experts consulted by CDA-AMC (Table 11).

Table 11: Subgroup Analyses for the DREAMM-7 Trial

R-ISS = Revised International Staging System; vs. = versus.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Duration of Response

For the primary analysis of DoR, all patients were included in the analysis, regardless of response status, to allow for a valid statistical comparison between the 2 groups. The response was based on an IRC assessment per IMWG criteria.²⁵ The DoR was analyzed based on the RMDoR, which is the difference between the KM curves of PFS and response or PFS for a treatment group, using a nonparametric approach.⁷⁰ Using this approach, nonresponders had an observed DoR of 0. The approach accounts for time to response (TTR), ORR, and DoR where the summary measure is the time from response to progression or death.²⁶ The comparison of RMDoR between the 2 treatment groups was based on a 1-sided z test at the overall significance level of 2.5%. The RMDoR and the corresponding 95% CI were calculated for each group.²⁵

The following sensitivity analyses for DoR were performed at the time of the primary analysis:

• DoR based on RMDoR analysis, using an investigator-assessed response

• DoR based on conventional DoR analysis in responders, using an IRC-assessed response, but among patients who achieved a response (confirmed PR or better).

As an exploratory analysis, a conventional DoR analysis was performed, where responders without disease progression were censored at the censoring time point for TTP; however, death due to causes other than disease progression was handled in the same way as death due to disease progression. The distribution of DoR was summarized using the KM method by treatment group. The median and corresponding 95% CIs were estimated using the Brookmeyer-Crowley method.

MRD Negativity

The number and proportion of patients who were MRD-negative at the primary PFS analysis were summarized by treatment groups. The corresponding exact 95% CI for the MRD negativity rate and associated P value were provided. The P value was obtained using the Cochran-Mantel-Haenszel test, stratified by 2 randomization factors (the number of prior lines of therapy, and prior bortezomib use) at a 1-sided significance level of 0.025. A supportive P value was calculated as well from the Fisher exact test.

The following sensitivity analyses for MRD negativity were performed at the time of the primary analysis:

• MRD based on the modified ITT analysis set using investigator-assessed responses

• MRD based on participants with VGPR or better, using both IRC and investigator-assessed responses

• MRD based on stratification data from the clinical database

• based on the presence of mis-stratification for stratification factors used for randomization, a supplementary analysis was performed using the stratification data based on the clinical database.

CRR and ORR

The number and percentage of patients for each of the following categories were summarized by treatment group: sCR, CR, VGPR, PR, ORR, minimal response, stable disease, and disease progression. Patients with unknown or missing responses were considered nonresponders. The earliest date of the 2 consecutive assessments was used as the date of the confirmed response.

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

Descriptive summaries (mean, SD, median, minimum, and maximum) of both the actual value and change from baseline at selected time points were provided for EORTC QLQ-C30 domain and symptom scores, including GHS or QoL. Time points included, but were not limited to, worst case postbaseline, end of treatment, and last follow-up visit. Longitudinal changes from baseline by treatment group for selected EORTC QLQ-C30 domain and symptom scores (fatigue, physical functioning, role functioning, and GHS or QoL) were analyzed using a restricted maximum likelihood-based mixed model of repeated measures. This model compared between-treatment difference while adjusting for correlations across multiple time points within each patient, controlling for the baseline variables. The mixed model of repeated measures included patient, treatment (as a random effect), analysis visit, and treatment-by-visit interaction (as a fixed effect) as explanatory variables, with the baseline value as a covariate along with the baseline-by-visit interaction. For scales requiring a single-item measure, if the item was missing, the score was considered missing. For scales requiring multiple items, if at least half of the items from the scale were available, the score was calculated based on the available items; if more than half of the items from the scale were missing, the score was set to missing.

Analysis Populations

The analysis populations of the DREAMM-7 trial are provided in Table 12. The efficacy outcomes, including HRQoL, were analyzed based on the ITT population. The safety outcomes were analyzed using the safety analysis set, defined as patients who received at least 1 dose of study treatment.

Table 12: Analysis Populations of the DREAMM-7 Trial

ICF = informed consent form; ITT = intention to treat; mITT = modified intention to treat.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results

Patient Disposition

Patient disposition for the DREAMM-7 trial is summarized in Table 13. A total of 129 of the 623 (20.7%) patients screened in the DREAMM-7 trial did not advance past screening, primarily due to an inability to meet eligibility criteria, including not meeting measurable disease parameters (8%) or adequate organ system functions (33%), or having corneal epithelial disease (12%) or hepatitis B (19%). In total, 243 patients were randomized to the BVd group and 251 patients to the DVd group. Two patients were initially randomized to the DVd group but were not treated; these patients were rescreened and rerandomized to the BVd and DVd groups within a short time frame. Because these patients had multiple randomizations and baseline datasets, the recommended approach to preserve the balance in prognostic factors achieved by randomization was to retain both sets of baseline data and randomizations, counting them as 4 unique patients.

At the time of IA1, using the October 2, 2023, data cut-off, the median duration of follow-up was 29.2 months (range, 0.2 month to 40.0 months) in the BVd group and 27.6 months (range, 0.1 month to 39.7 months) in the DVd group. A total of 161 (66%) patients in the BVd group and 195 (78%) patients in the DVd group discontinued study treatment. The most common reasons for treatment discontinuation were disease progression (36.6% and 75.9% in the BVd and DVd groups, respectively), AEs (28.0% and 11.3% in the BVd and DVd groups, respectively), and physician decision (20.5% and 5.1% in the BVd and DVd groups, respectively). A higher proportion of deaths was reported in the DVd group (54%) compared to the BVd group (21%). Additionally, a greater proportion of patients in the BVd group was still on study treatment at the data cut-off date (33%) compared to the DVd group (20%).

At the time of IA2, using the October 7, 2024, data cut-off, the median duration of follow-up was 40.2 months (range, 0.2 month to 52.3 months) in the BVd group and 38.2 months (range, 0.1 month to 51.3 months) in the DVd group. The sponsor did not report the reasons for the study withdrawal. A total of 182 (75%) patients in the BVd group and 208 (83%) patients in the DVd group discontinued study treatment. The most common reasons for treatment discontinuation were disease progression (30% and 64% in the BVd and DVd groups, respectively) and AEs (21% and 9% in the BVd and DVd groups, respectively), followed by physician decision (13% and 4% in the BVd and DVd groups, respectively). A greater proportion of patients in the BVd group was still on study treatment at the data cut-off date (25%) compared to the DVd group (15%).

Table 13: Summary of Patient Disposition From the DREAMM-7 Trial

BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; IA1 = interim analysis 1; IA2 = interim analysis 2; ITT = intention to treat; mITT = modified intention to treat; NR = not reported.

Note: Patients could have more than 1 reason for not advancing past screening; hence, the percentage might sum to more than 100%. Patients could have only 1 primary reason for withdrawal. Two participants were randomized, not treated, rescreened, and rerandomized.

^aOctober 2, 2023, data cut-off.

^bOctober 7, 2024, data cut-off.

^cThe duration of follow-up was defined as the time from randomization to last contact or death.

Sources: Clinical Study Report for the DREAMM-7 study (2024)²⁶ and Clinical Study Report Addendum (Interim Analysis 2) for the DREAMM-7 study (2025).²⁷ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Baseline Characteristics

The baseline characteristics in the ITT population outlined in Table 14 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results. In general, patient demographic and disease characteristics were well balanced between the BVd and DVd treatment groups. The mean age of patients was 64.5 (SD = 9.5) years in the BVd group and 63.6 (SD = 10.1) years in the DVd group. Half of the patients were aged younger than 65 years (50% in both groups). Most patients had an ECOG PS score of 0 or 1 (96% in both groups). The majority of patients had an R-ISS score of stage I or stage II (95% and 94% in the BVd and DVd groups, respectively). A high proportion of patients had refractory or relapsed disease (97% and 98% in the BVd and DVd groups, respectively). Of the 494 patients, 250 (51%) patients had received 1 prior line of therapy, 117 (24%) patients had received 2 prior lines of therapy, and 70 (14%) patients had received 3 prior lines of therapy. Additionally, 257 (52%) patients had prior exposure to lenalidomide and 166 (34%) patients had disease refractory to lenalidomide. Immunoglobulin G was the most common myeloma immunoglobulin (66% and 63% in the BVd and DVd groups, respectively). Prior stem cell transplants were reported in 164 (67%) patients in the BVd group and 173 (69%) patients in the DVd group. High-risk cytogenetic abnormalities were present in 67 (28%) patients in the BVd group and 69 (27%) patients in the DVd group, with t(4;14) and 17p13del being the most common high-risk cytogenetic abnormalities in both groups.

Table 14: Summary of Baseline Characteristics From the DREAMM-7 Trial — ITT Population

BMI = body mass index; BVd = belantamab mafodotin plus bortezomib and dexamethasone; CRF = case report form; DVd = daratumumab plus bortezomib and dexamethasone; ECOG PS = Eastern Cooperative Oncology Group Performance Status; eCRF = electronic case report form; IgA = immunoglobulin A; IgD = immunoglobulin D; IgE = immunoglobulin E; IgG = immunoglobulin G; IgM = immunoglobulin M; ITT = intention to treat; PD = progressive disease; R-ISS = Revised International Staging System; R/R = relapsed/refractory; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aAge was imputed when the full date of birth was not provided.

^bCollected in the CRF as follows: PD after more than 60 days of stopping treatment (includes the eCRF categories of relapsed < 6 months, relapsed at 6 months to 12 months, and relapsed > 12 months).

^cCollected in the CRF as follows: PD on treatment or within 60 days of stopping treatment (includes the eCRF categories of relapsed < 60 days, refractory, primary refractory, relapsed, and refractory).

^dIncludes the CRF category of relapsed (time period unknown).

^eParticipants could be included in more than 1 category.

^fOnly positive results were summarized.

^gIf the patients had at least 1 high-risk abnormality: t(4;14), t(14;16), or 17p13del.

^hIf the patients had negative results for all high-risk abnormalities: t(4;14), t(14;16), or 17p13del.

ⁱTime to relapse was defined as the time between the start date of the first line of therapy to the PD date on first-line treatment. If no PD date was available, the start date of the second line of treatment was used. If no PD date or start date of the second line of treatment was available, the date of randomization into the study was used.

^jAnalyzed using the safety population.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Exposure to Study Treatments

At the time of IA1, using the October 2, 2023, data cut-off, the median number of treatment cycles in the BVd group was 12.0 cycles (range, 1 cycle to 55 cycles) compared to 15.0 cycles (range, 1 cycle to 46 cycles) in the DVd group (Table 15). A longer median duration of exposure was observed for patients receiving BVd compared to those receiving DVd (15.9 months [range, 0.7 month to 40.2 cycles] versus 12.9 months [range, 0.2 month to 40.5 months], respectively).

At the time of IA2, using the October 7, 2024, data cut-off, the median number of treatment cycles in the BVd group was 13.0 cycles (range, 1 cycle to 69 cycles) compared to 15.0 cycles (range, 1 cycle to 55 cycles) in the DVd group. A longer median duration of exposure was observed for patients receiving BVd compared to those receiving DVd (15.9 months [range, 0.7 month to 52.3 months] versus 12.8 months [range, 0.2 month to 48.8 months], respectively).

At the time of IA1, using the October 2, 2023, data cut-off, patients in the BVd group had received a median of 9 treatment cycles of belantamab (range, 1 cycle to 55 cycles), with the overall median dose intensity being 1.27 mg/kg per cycle (range, 0.2 mg/kg per cycle to 2.6 mg/kg per cycle) once every 3 weeks, and the median relative dose intensity being 50.9% (Table 16). As a result, the average treatment dose and administration schedule of belantamab within the BVd regimen were both lower and less frequent than those specified in the trial’s per-protocol dosing schedule. This was primarily due to necessary dose adjustments and extended dosing intervals for most patients to manage ocular AEs.

At the time of IA2, using the October 7, 2024, data cut-off, patients in the BVd group had received a median of 9 treatment cycles of belantamab (range, 1 cycle to 69 cycles), with the overall median dose intensity being 1.22 mg/kg per cycle (range, 0.2 mg/kg per cycle to 2.6 mg/kg per cycle) once every 3 weeks.

At the time of IA1, using the October 2, 2023, data cut-off, patients in the DVd group had received a median of 15 treatment cycles (range, 1 cycle to 46 cycles) of daratumumab. The overall median dose intensity of daratumumab was 45.5 mg/kg per cycle (range, 13.1 mg/kg per cycle to 52.2 mg/kg per cycle) for cycle 1 to cycle 3, 16.0 mg/kg per cycle (range, 2.5 mg/kg per cycle to 19.2 mg/kg per cycle) for cycle 4 to cycle 8, and 15.8 mg/kg per cycle (range, 12.0 mg/kg per cycle to 19.2 mg/kg per cycle) for cycles 9 and beyond (Table 17). At the time of IA2, using the October 7, 2024, data cut-off, patients in the BVd group had received a median of 15 treatment cycles of belantamab (range, 1 cycle to 55 cycles).

At the time of IA1, using the October 2, 2023, data cut-off, exposure to bortezomib (Table 18) and dexamethasone (Table 19) was similar in patients in the BVd and DVd groups.

Table 15: Summary of Patient Exposure, Both Combination Therapy and Monotherapy Phases of Treatment in the DREAMM-7 Trial — Safety Population

AE = adverse event; BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; IA1 = interim analysis 1; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aThe number of cycles includes all cycles during which at least 1 component was administered. The number of cycles represents the number of doses; however, the number of cycles does not include dose delays (such as those due to ocular AEs for belantamab mafodotin).

^bTotal duration of exposure = [(overall study treatment end date) – (overall study treatment start date) + 1].

^cTotal duration of exposure (based on the date of the decision to discontinue treatment) = [(latest of [overall study treatment end date, latest date of decision to discontinue treatment]) – (overall study treatment start date) + 1].

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 16: Summary of Exposure to Belantamab Mafodotin in the DREAMM-7 Trial — Safety Population

BVd = belantamab mafodotin plus bortezomib and dexamethasone; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aCycles represent doses. Cycles during which no dose was administered due to dose delays were not counted.

^bOverall dose intensity was the cumulative actual dose divided by (duration of exposure in days divided by planned cycle length).

^cRelative dose intensity was calculated as a percentage and was defined as 100 multiplied by (overall dose intensity divided by planned dose intensity).

Sources: Clinical Study Report for DREAMM-7, 2024.²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 17: Summary of Exposure to Daratumumab in the DREAMM-7 Trial — Safety Population

DVd = daratumumab plus bortezomib and dexamethasone; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aDose intensity was the cumulative actual dose divided by (the duration of exposure in days divided by planned cycle length). This was only calculated for daratumumab due to changes in the dosing regimen at cycle 4 and an increase in cycle length to 28 days at cycle 9.

^bRelative dose intensity was calculated as a percentage and was defined as 100 multiplied by (dose intensity divided by planned dose intensity, within the indicated cycles).

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 18: Summary of Exposure to Bortezomib in the DREAMM-7 Trial — Safety Population

BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aOverall dose intensity was the cumulative actual dose divided by (the duration of exposure in days divided by planned cycle length).

^bRelative dose intensity was calculated as a percentage and was defined as 100 multiplied by (overall dose intensity divided by planned dose intensity).

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 19: Summary of Exposure to Dexamethasone in the DREAMM-7 Trial — Safety Population

BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; SD = standard deviation.

Note: Data cut-off date of October 2, 2023.

^aOverall dose intensity was the cumulative actual dose divided by (duration of exposure in days divided by planned cycle length).

^bRelative dose intensity was calculated as a percentage and was defined as 100 multiplied by (overall dose intensity divided by planned dose intensity).

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Concomitant Medications

At the time of IA1, using the October 2, 2023, data cut-off, concomitant medications in the DREAMM-7 trial were administered to 242 (99.6%) patients in the BVd group and 245 (98%) patients in the DVd group. A higher proportion of patients in the BVd group had received blood products compared to those in the DVd group (28% versus 16%, respectively). Additionally, a greater proportion of patients in the BVd group had received platelet transfusions compared to the DVd group (19% versus 9%, respectively). The use of blood-supportive care products was similar between the 2 treatment groups.

Subsequent Antimyeloma Therapy

At the time of IA1, using the October 2, 2023, data cut-off, subsequent antimyeloma therapy was reported for 62 (26%) patients in the BVd group and 110 (44%) patients in the DVd group (Table 20). The median time from study treatment discontinuation to the initiation of subsequent antimyeloma therapy was longer in the BVd group compared with the DVd group (69 days [range, 4 days to 698 days] versus 44.0 days [range, 2 days to 610 days], respectively). At the time of IA2, using the October 7, 2024, data cut-off, subsequent antimyeloma therapy was reported for 87 (36%) patients in the BVd group and 130 (52%) patients in the DVd group.⁷¹ Given that both the IA1 and IA2 data cut-offs on October 2, 2023, and October 7, 2024, respectively, considering any line of subsequent therapy, a higher proportion of patients in the DVd group initiated steroids, immunomodulators, and proteasome inhibitors compared to the BVd group. Conversely, more patients in the BVd group initiated monoclonal antibody therapy compared to the DVd group.⁷¹

Table 20: Summary of Subsequent Treatment From the DREAMM-7 Trial — ITT Population

BiTE = Bispecific T Cell Engager; BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; IA1 = interim analysis 1; IA2 = interim analysis 2; ITT = intention to treat; NR = not reported; T/NK = T-cell/natural killer.

Note: Drug class contains all medications taken for the specific follow-up anticancer line of therapy. Drug class percentages are based on the number of participants who received the subsequent line of therapy number.

Sources: Clinical Study Report for the DREAMM-7 study (2024)²⁶ and Hungria et al. (2024).⁷¹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Dose Modifications

At the time of IA1, using the October 2, 2023, data cut-off, dose delays were reported in 88% of patients in the BVd group and 72% of patients in the DVd group. The median duration of a dose delay was 54 days (range, 1 day to 732 days) for belantamab compared to 5 days (range, 1 day to 180 days) for daratumumab. Dose interruptions were reported in 1% of patients receiving belantamab and 22% of those receiving daratumumab, with a total of 73 dose interruptions reported for daratumumab. Of these, 55% were due to AEs. A total of 194 dose reductions were reported for belantamab, with 49% of these reductions attributed to corneal exam findings and 41% due to AEs. Nearly all dose reductions occurred within the first 9 cycles.

Protocol Deviations

The proportion of patients with any important protocol deviation was higher in the BVd group compared with the DVd group (78% versus 63%, respectively). The most common types of protocol deviation in both treatment groups were related to assessment or time point completion (44% versus 37% in the BVd and DVd groups, respectively), wrong study treatment or administration of dose (39% versus 14% in the BVd and DVd groups, respectively), informed consent (19% versus 20% in the BVd and DVd groups, respectively), and study procedures (16% versus 14% in the BVd and DVd groups, respectively).

Efficacy

Only those efficacy outcomes and analyses of subgroups identified as important to this review are reported. Findings for key efficacy outcomes in the DREAMM-7 trial are summarized in Table 21.

Primary Outcome

PFS per IRC Assessment

At the time of IA1, using the October 2, 2023, data cut-off, the median duration of follow-up was 28.2 months. In the ITT population, PFS events were reported for 91 (37%) patients in the BVd group and 158 (63%) patients in the DVd group. The median PFS was 36.6 months (95% CI, 28.4 months to not estimable) in the BVd group and 13.4 months (95% CI, 11.1 months to 17.5 months) in the DVd group, with a between-group HR of 0.41 (95% CI, 0.31 to 0.53; P < 0.00001) in favour of the BVd group (Figure 5). The KM-estimated probabilities of being alive or progression-free at 12 months and 18 months were 78.3% (95% CI, 72.2% to 83.2%) versus 53.3% (95% CI, 46.5% to 59.5%), and 68.8% (95% CI, 62.0% to 74.7%) versus 42.8% (95% CI, 36.2% to 49.2%), in the BVd and DVd groups, respectively. The proportional hazards assumption for PFS was tested, using a diagnostics plot (log of time versus log of negative log of estimated survival), along with a plot of Schoenfeld residuals. These tests indicated no significant interaction between time and treatment (P value = 0.9408), providing insufficient evidence to suggest a violation of the assumption.

PFS reached the predefined boundary for statistical significance at IA1 and, in accordance with the study protocol, was evaluated as an exploratory end point at the time of IA2. At the time of IA2, using the October 7, 2024, data cut-off, the median PFS was ████ ██████ ████ ███ ████ ██████ ██ ████ ███████ in the BVd group and ████ ██████ ████ ███ ████ ██████ ██ ████ ███████ in the DVd group, with a between-group ██ ██ ████ █████ ██ ████) in favour of the BVd group. The KM-estimated probabilities of being alive and progression-free at 24 months and 36 months were █████ ████ ███ █████ ██ ██████ versus █████ ████ ███ █████ ██ ███████ and █████ ████ ███ █████ ██ ██████ versus █████ ████ ███ █████ ██ ███████ in the BVd and DVd groups, respectively.

The results of predefined exploratory subgroup analyses at IA1 were consistent with the primary PFS analyses, including for those exposed or refractory to lenalidomide (HR = 0.33 [95% CI, 0.23 to 0.48] and HR = 0.37 [95% CI, 0.24 to 0.56], respectively), those with more than 1 prior line of therapy (HR = 0.33 [95% CI, 0.23 to 0.48]), those with high cytogenetic risk (HR = 0.36 [95% CI, 0.22 to 0.58]) and R-ISS stage II or stage III (HR = 0.45 [95% CI, 0.32 to 0.64]).

The results of all sensitivity analyses at IA1 were consistent with the primary PFS analyses. The results of the PFS analyses per IRC assessment were consistent with those from the investigator assessment.

Figure 5: KM Plot of PFS at IA1 per IRC Assessment in the DREAMM-7 Trial — ITT Population

BOR = bortezomib; DARA = daratumumab; DEX = dexamethasone; GSK916 = belantamab mafodotin plus bortezomib and dexamethasone; IA1 = interim analysis 1; IRC = independent review committee; ITT = intention to treat; KM = Kaplan-Meier; PFS = progression-free survival.

Notes: There were 2 participants in the ITT population who were randomized, not treated, rescreened, and rerandomized. They were counted as 4 unique participants in Figure 5.

The data cut-off date was October 2, 2023.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Key Secondary Outcomes

Overall Survival

At the time of IA1, using the October 2, 2023, data cut-off, OS data had reached 29% (141 of 494 patients) overall maturity, and the median OS was not reached in either treatment group. Death was reported in 54 (22%) patients in the BVd group and 87 (35%) patients in the DVd group, with a between-group HR of 0.57 (95% CI, 0.40 to 0.80; P value = 0.00049). At the time of IA1, the results for OS did not meet the significance criterion. The KM-estimated probabilities of being alive at 12 months and 18 months were 86.5% (95% CI, 81.5% to 90.3%) versus 80.7% (95% CI, 75.2% to 85.1%), and 83.9% (95% CI, 78.6% to 88.0%) versus 72.6% (95% CI, 66.5% to 77.8%), in the BVd and DVd groups, respectively. The subgroup analyses for OS were not performed due to the immaturity of data at the time of analysis.

At the time of IA2, using the October 7, 2024, data cut-off, the median OS had not been reached in either treatment group. Death was reported in 68 (28%) patients in the BVd group and 103 (41%) patients in the DVd group, with a between-group HR of 0.58 (95% CI, 0.43 to 0.79; P value = 0.00023) in favour of the BVd group. At the time of IA2, OS data had reached 34.6% (171 of 494patients) overall maturity, and the significance criterion was met (P < 0.00112). The KM-estimated probabilities of being alive at 24 months and 36 months were 79.0% (95% CI, 73.2% to 83.7%) versus 67.4% (95% CI, 61.0% to 73.0%), and 71.4% (95% CI, 67.9% to 79.2%) versus 60.2% (95% CI, 53.6% to 66.2%), in the BVd and DVd groups, respectively (Figure 6).

Figure 6: KM Plot of OS at IA2 per IRC Assessment in the DREAMM-7 Trial — ITT Population

BOR = bortezomib; DARA = daratumumab; DEX = dexamethasone; GSK916 = belantamab mafodotin plus bortezomib plus dexamethasone; IA2 = interim analysis 2; IRC = independent review committee; ITT = intention to treat; KM = Kaplan-Meier; OS = overall survival.

Notes: There were 2 participants in the ITT population who were randomized, not treated, rescreened, and rerandomized. They were counted as 4 unique participants in Figure 6.

The data cut-off date was October 7, 2024.

Sources: Clinical Study Report for the DREAMM-7 study (2024)²⁶ and (2025). ²⁷

Duration of Response

The RMDoR is a composite end point that integrates overall response data and PFS data, allowing cross-group comparison, and efficiently assesses the treatment effect related to tumour reductions. At the time of IA1, using the October 2, 2023, data cut-off, the RMDoR was 19.0 months (95% CI, 17.7 months to 20.4 months) in the BVd group and 13.2 months (95% CI, 11.8 months to 14.6 months) in the DVd group, with a ratio of RMDoR of 1███ ████ ███ ████ ██ █████ P value < 0.00001) in favour of the BVd group.

A conventional assessment of DoR was conducted among patients with PR or better, and was defined as the time from the first documented evidence of PR or better until disease progression or death due to any cause. At the time of IA1, using the October 2, 2023, data cut-off, the median DoR was 35.6 months (95% CI, 30.5 months to not estimable) in the BVd group and 17.8 months (95% CI, 13.8 months to 23.6 months) in the DVd group.

At the time of IA2, using the October 7, 2024, data cut-off, the median DoR was 40.8 months (95% CI, 30.5 months to not estimable) in the BVd group and 17.8 months (95% CI, 13.8 months to 23.6 months) in the DVd group. The KM-estimated probabilities of maintaining response at 12 months and 18 months were 83.3% (95% CI, 77.1% to 87.9%) versus 61.2% (95% CI, 53.4% to 68.0%), and 75.6% (95% CI, 68.6% to 81.3%) versus 49.1% (95% CI, 41.3% to 56.5%), in the BVd and DVd groups, respectively (Figure 7).

The sensitivity analysis for DoR based on the IRC assessment, which considered death due to disease progression only, was consistent with the conventional DoR analysis that included death from any cause.

Figure 7: KM Plot of DoR per IRC Assessment in the DREAMM-7 Trial — ITT Population

BOR = bortezomib; DARA = daratumumab; DEX = dexamethasone; DoR = duration of response; GSK916 = belantamab mafodotin plus bortezomib and dexamethasone; IRC = independent review committee; ITT = intention to treat; KM = Kaplan-Meier.

Notes: There were 2 participants in the ITT population who were randomized, not treated, rescreened, and rerandomized. They were counted as 4 unique participants in Figure 7.

The data cut-off date was October 2, 2023.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Minimal Residual Disease

At the time of IA1, using the October 2, 2023, data cut-off, the MRD negativity analysis was exploratory in nature because the OS results were not significant at the time of IA1. The proportion of patients who achieved MRD negativity by best response (CR or sCR) was higher in the BVd group compared with the DVd group at 24.7% versus 9.6%, respectively (P < 0.00001). MRD negativity by best response based on an IRC assessment showed higher MRD negativity rates in the BVd group compared with the DVd group across all response categories. Because the IA2 analysis of OS reached statistical significance, the MRD negativity results from the IA1 primary analysis can be considered statistically significant due to the prespecified hierarchical testing of outcomes. At the time of IA2, using the October 7, 2024, data cut-off, MRD negativity rates by CR or sCR remained consistent with the IA1 results. The BVd group showed higher MRD negativity rates (25.1%) compared with the DVd group (10.4%).

Secondary Outcomes

Complete Response Rate

At the time of IA1, using the October 2, 2023, data cut-off, a greater proportion of patients in the BVd group (34.6%) compared with the DVd group (17.1%) had achieved CR or sCR, with a between-group difference of 17.4% (95% CI, 8.6% to 26.1%).

Overall Response Rate

At the time of IA1, using the October 2, 2023, data cut-off, a greater proportion of patients in the BVd group (82.7%) compared with the DVd group (71.3%) had achieved confirmed PR or better, with a between-group difference of 11.4% (95% CI, 2.6% to 20.1%).

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

EORTC QLQ-C30 data reported in this section are from IA1, with the data cut-off date of October 2, 2023.

GHS Domain

The EORTC QLQ-C30 GHS domain remained stable across both treatment groups over time (Figure 8). Both groups showed a slight mean deterioration in GHS scores between week 4 and week 43, after which scores stabilized and remained consistent throughout the treatment period. Between week 43 and week 100, 24% to 35% of patients in the BVd group experienced an improvement in GHS or QoL scores (a 10-point increase or higher). Small sample sizes at later time points limited the ability to interpret the results (Figure 8).

For example, at week 43, in the GHS score, patients in the DVd group had experienced a mean change of ███ ██████ from baseline compared with a change of ███ ██████ in the DVd group, with a least squares mean difference between groups of ███ ████ ███ ████ ██ ████. At week 121, in the GHS score, patients in the DVd group had experienced a mean change of ███ ██████ from baseline compared with a change of ████ ██████ in the DVd group, with a least squares mean difference between groups ███ ████ ███ ████ ██ ████.

Figure 8: Change From Baseline in EORTC QLQ-C30 Global Health Status and QoL Domain Score — ITT Population

BOR = bortezomib; BSLN = baseline, chg = change; DARA = daratumumab; DEX = dexamethasone; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EOT = end of treatment; GSK916 = belantamab mafodotin plus bortezomib and dexamethasone; ITT = intention to treat; LFU = last follow-up; QoL = quality of life; WCPB = worst case postbaseline; WK = week.

Notes: There were 2 participants in the ITT population who were randomized, not treated, rescreened, and rerandomized. They were counted as 4 unique participants in Figure 8.

The data cut-off date was October 2, 2023.

Source: Clinical Study Report for the DREAMM-7 study (2024).²⁶

Table 21: Summary of Key Efficacy Results of the DREAM-7 Trial — ITT Population

BVd = belantamab mafodotin plus bortezomib and dexamethasone; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; CR = complete response; CRR = complete response rate; DoR = duration of response; DVd = daratumumab plus bortezomib and dexamethasone; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HR = hazard ratio; IA1 = interim analysis 1; IA2 = interim analysis 2; IRC = independent review committee; ITT = intention to treat; LSM = least squares mean; KM = Kaplan-Meier; MRD = minimal residual disease; NA = not available; NE = not estimable; NR = not reported; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PR = partial response; R-ISS = Revised International Staging System; RMDoR = restricted mean duration of response; RMST = restricted mean survival time; sCR = stringent complete response; SD = standard deviation; SE = standard error; t* = common truncation time.

Notes: There were 2 participants in the ITT population who were randomized, not treated, rescreened, and rerandomized. They were counted as 4 unique participants.

The data cut-off dates were October 2, 2023, for IA1 and October 7, 2024, for IA2.

^aCIs were estimated using the Brookmeyer-Crowley method.

^bHRs were estimated using a Cox proportional hazards model stratified by the number of lines of prior therapy (1 versus 2 or 3 versus ≥ 4), prior bortezomib use (no or yes), and R-ISS at screening (stage I versus stage II or stage III), with a covariate of treatment.

^cThe P value was from a 1-sided stratified log-rank test.

^dT* was calculated using the algorithm defined in Huang and Tian.⁷⁰

^eRMDoR for a treatment group was the difference between the KM curves of PFS, and response or PFS. The comparison of RMDoR between the 2 treatment groups was based on a 1-sided z test at the overall 2.5% level of significance.

^fThe 1-sided P value was from a mean DoR test.

^gThe P value was nominal, derived from a 1-sided stratified log-rank test.

^hThe P value was from a 1-sided stratified log-rank test. At 171 actual events (a 48.2% OS information fraction), OS was declared significant if the P value was less than 0.00112.

ⁱThe MRD negativity rate was compared between treatment groups using an unadjusted Fisher exact test. CIs were based on the exact method. Patients without an MRD assessment were assumed to be nonnegative.

^jCIs were based on the exact method. The CIs were not adjusted for multiplicity and cannot be used in place of hypothesis testing.

^kVisits were included in the model only if the number of patients with analyzable data at the current time point was at least 10 in both treatment groups.

^lAnalysis was performed using a mixed model of repeated measures with covariates of treatment group, visit, baseline value, treatment-by-visit interactions, and baseline-value-by-visit interactions.

^mWithin-group probabilities were estimated using the KM method, as specified in the statistical analysis plan. The between-group estimates were calculated using the KM method at the request of CDA-AMC.

Sources: Clinical Study Report for the DREAMM-7 study (2024) ²⁶ and Clinical Study Report Addendum (Interim Analysis 2) for the DREAMM-7 study (2025).²⁷ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Harms

A summary of harms reported in the safety population in the DREAMM-7 trial is provided in Table 22.

Adverse Events

In the DREAMM-7 trial, all patients in both groups experienced at least 1 AE. The most common AEs of any grade in the BVd group included thrombocytopenia, blurred vision, dry eye, photophobia, foreign body sensation in eyes, and eye irritation. At the time of IA1, 95% of patients in the BVd group and 78% of patients in the DVd group had experienced at least 1 AE of grade 3 or higher. The most common AEs of grade 3 or higher in the BVd group were thrombocytopenia (55% in the BVd group versus 35% in the DVD group), blurred vision (22% in the BVd group versus less than 1% in the DVD group), decreased platelet count (18% in the BVd group versus 11% in the DVD group), neutropenia (12% in the BVd group versus 6% in the DVD group), and pneumonia (12% in the BVd group versus 4% in the DVD group). The population with an AE of grade 3 or higher was not reported at IA2.

Serious Adverse Events

At the time of IA1, a higher proportion of patients in the BVd group had experienced at least 1 SAE compared to those in the DVd group (50% versus 37%, respectively). The most common SAEs in the BVd and DVd groups were pneumonia (11% versus 4%, respectively), COVID-19 (5% versus 4%, respectively), pyrexia (5% versus 4%, respectively), and COVID-19 pneumonia (93% in both groups). The IA2 data on SAEs were not available at the time of submission to CDA-AMC.

Withdrawals Due to Adverse Events

At the time of IA1 and IA2, a higher proportion of patients in the BVd group had discontinued treatment due to AEs compared to those in the DVd group (31% versus 19%, and 32% versus 19%, respectively). At the time of IA2, the most common reason for discontinuation was peripheral sensory neuropathy (6%) in the BVd group and peripheral neuropathy (4%) in the DVd group. At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced a dose reduction compared to those in the DVd group (75% versus 59%, respectively). The most common reasons for dose reduction in the BVd group were thrombocytopenia, peripheral sensory neuropathy, blurred vision, and peripheral neuropathy. At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced a dose interruption or delay of study treatment compared to the DVd group (94% versus 75%, and 95% versus 76%, respectively). The most common reasons for dose interruption or delay in the BVd group were thrombocytopenia, blurred vision, COVID-19, dry eye, photophobia, and eye irritation.

Mortality

At the time of IA1, deaths were reported in 21% of patients in the BVd group and in 31% of patients in the DVd group. The majority of deaths in both groups was attributed to cancer, with 7% in the BVd group and 19% in the DVd group. Fatal SAEs were reported in 10% of patients in the BVd group and 8% of patients in the DVd group. At the time of IA2, using the October 7, 2024, data cut-off, deaths were reported in 29% of patients in the BVd group and in 41% of patients in the DVd group.

Notable Harms

At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced at least 1 ocular AE of any grade compared to the DVd group (79% versus 29%, ███ ███ ██████ ████ respectively). Blurred vision and dry eye were reported in more than half of the patients in this treatment group. Ocular AEs in the BVd group were more severe, with ███ classified as grade 3 or grade 4 while the majority in the DVd group were classified as grade 1 (███). At the time of IA2, treatment discontinuations due to ocular AEs were reported in ██ of patients in the BVd group compared to none in the DVd group. Treatment modifications due to corneal AEs were more common in the BVd group, with 52% of patients experiencing ocular AEs having doses interrupted, delayed, or reduced compared to only 2% in the DVd group.

At the time of IA1, investigator-assessed corneal events and BCVA events, as evaluated by the KVA scale, had been reported exclusively in patients in the BVd group. At the time of IA1, 201 (83%) patients in the BVd group had an incidence of corneal events (overall KVA grade). The severity of the corneal events was high, with ███ of the 201 patients classified as grade 3 and ███ as grade 4. Most corneal events were managed through dose modifications, with 89% of affected patients having doses interrupted or delayed, and 31% of affected patients experiencing dose reductions. The incidence of BCVA events, as evaluated by the KVA scale, was ███ in the BVd group, with the majority of findings classified as grade 2. Most patients required a dose interruption or delay due to a BCVA event while only a few patients needed to discontinue study treatment. Specifically, 87% of patients experienced a dose interruption or delay while 6% of patients discontinued the study treatment. The population with investigator-assessed corneal and BCVA events was not reported at IA2.

At the time of IA1 and IA2, a higher proportion of patients in the BVd group had experienced thrombocytopenia compared to those in the DVd group (87% versus 65%, and 88% versus 65%, respectively). At the time of IA2, of the 212 patients in the BVd group and 160 patients in the DVd group who had experienced thrombocytopenia, grade 3 and grade 4 thrombocytopenia were more frequent in the BVd group (83%) compared to the DVd group (71%). Thrombocytopenia events of grade 4 were reported in 53% of patients receiving BVd versus 33% of patients receiving DVd. In the DREAMM-7 trial, ███ of patients in the BVd group and ██ in the DVd group had received platelet transfusions. At the time of IA2, despite the high incidence, few patients ███ ██ █████ in either group had experienced SAEs related to thrombocytopenia, and treatment discontinuation rates due to thrombocytopenia were low (██ in the BVd group and ██ in the DVD group). Dose reductions and interruptions or delays were more common in the BVd group compared to the DVd group (43% versus 20%, and 53% versus 39%, respectively).

At the time of IA1, neutropenia was reported in ███ of patients in the BVd group and ███ of patients in the DVd group. Among patients who had experienced neutropenia AEs, most in both the BVd and DVd groups had grade 3 events (███ ██████ ████ respectively). Neutropenia events of grade 4 were more common in the BVd group (███) compared to the DVd group (███). The neutropenia events were manageable with dose modifications (including dose reductions and delays or interruptions), and no patients in either treatment group discontinued study treatment due to neutropenia events. The population with neutropenia events was not reported at IA2. At the time of IA1, a higher proportion of patients in the DVd group had experienced infusion-related reactions (███) compared to the BVd group (██), with more than ███ of patients in both treatment groups having grade 2 events.

Table 22: Summary of Harms Results From the DREAMM-7 Trial — Safety Population

AE = adverse event; BCVA = best corrected visual acuity; BVd = belantamab mafodotin plus bortezomib and dexamethasone; DVd = daratumumab plus bortezomib and dexamethasone; eCRF = electronic case report form; GSK = GlaxoSmithKline; IA1 = interim analysis 1; IA2 = interim analysis 2; KVA = Keratopathy and Visual Acuity; MM = multiple myeloma; NR = not reported; SAE = serious adverse event.

Note: Data cut-off date of October 2, 2023.

^aThere were 3 patients with an unknown primary cause of death. One patient in the DVd group died due to disease progression; however, the death was erroneously assigned to “Other noncardiovascular causes.”

^bOcular AEs are based on a hybrid of terms identified in the eCRF, and a list of terms identified by GSK internal review.

Sources: Clinical Study Report for the DREAMM-7 study (2024).²⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Critical Appraisal

Internal Validity

Randomization was performed using an appropriate methodology with adequate allocation concealment, and randomization stratification was prespecified. An interactive response methodology using a permuted block randomization scheme was used, and randomization with stratification was performed centrally, which typically has a low risk of bias. Overall, baseline patient and disease characteristics were balanced. The open-label design introduced a potential bias in the assessment of ORR, PFS, DoR, and CRR; however, this bias was mitigated by the use of a blinded IRC for disease assessment. Knowledge of the assigned treatment could have led to bias in the reporting and measurement of subjective outcomes, including patient-reported outcomes (e.g., HRQoL) and subjective AEs. However, the extent and direction of bias due to treatment knowledge is uncertain. A physician’s knowledge of their patient’s assigned treatment may influence how they manage the patient, while patients’ awareness of their assigned treatment may impact their likelihood of staying in the study. There was a moderately increased and imbalanced loss to follow-up across groups (10% of patients in the BVd group and 16% of patients in the DVd group). Reasons were similar across groups but were limited in their description (i.e., withdrawal by participant, physician decision, or loss to follow-up). This did not appear to have an important impact on the assessment of PFS and OS.

Based on the enrolled sample size, the study was powered to test its primary and key secondary end points. The analyses of primary and key secondary outcomes were conducted using the ITT population, which maintains randomization and minimizes the risk of bias by comparing groups with similar prognostic factors. Statistical analysis methods appear to be acceptable. Both interim analyses were planned a priori. A hierarchical testing strategy to sequentially test the primary and secondary outcomes was used to reduce the risk of type I error across these analyses. At the time of the first interim analysis, the weighted Bonferroni procedure was applied across OS and DoR, which is deemed conservative in controlling type I error. Results for the primary outcome were based on a prespecified interim analysis (249 PFS events), which occurred at an 89% information fraction, relative to the planned final PFS analysis (280 PFS events). The OS data were informed by a small information fraction (35% at IA2) at both interim analyses, and the median OS was not reached in either treatment group. Because the study outcomes are based on interim analyses, there is a risk that the effect of BVd compared with DVd is overestimated, particularly for OS; however, the existence and extent of any overestimation remains uncertain.^20-22 In addition, there is uncertainty as to whether the PFS benefits (as a surrogate outcome for OS) will translate into survival benefits. PFS reflects tumour growth and can be assessed before determining a survival benefit, without being influenced by subsequent therapies.⁴⁹ The US FDA considers the treatment effect measured by PFS as a potential surrogate end point to support accelerated drug approval.⁴⁹ However, data are insufficient to robustly evaluate the correlation between treatment effects on OS and PFS in clinical trials.^49,72 Given that the results at the data cut-off represent interim analyses for OS and are based on a limited information fraction, longer-term follow-up is needed to inform the true effect of BVd compared with DVd on survival.

Patients were allowed to receive posttreatment anticancer medications after discontinuing study treatment, which could influence the assessment of OS. A sensitivity analysis was planned to adjust for posttreatment anticancer medications, but the results were not reported. The subsequent treatments were not balanced between groups; this is expected because fewer patients progressed in the DVd group. There was a higher proportion of patients censored whose follow-up had ended in the analysis of PFS; the primary reason was the initiation of new antimyeloma therapy. Most of the censoring reasons were addressed in supplemental analyses that applied different censoring rules. The results for PFS were robust to these supplemental analyses and several sensitivity analyses. The results for the proportional hazards assumption were reported only for the PFS analysis, indicating insufficient evidence to suggest a violation of the assumption. Subgroup analyses were prespecified and conducted only for PFS; however, there was no multiplicity control for the subgroup analyses. Most subgroup results for PFS were consistent with the primary analysis, favouring BVd, although some of these analyses were limited by small sample sizes. DoR was analyzed based on the RMDoR, using a nonparametric approach, which was deemed appropriate.

Protocol deviations were reported in 78% to 63% of patients across the 2 treatment groups, and the proportion of protocol deviations was comparable between groups and identified before the database lock date. A large number of patients in the DREAMM-7 trial discontinued treatment, with fewer discontinuations (both from monotherapy and combination therapy) in the BVd group (66%) compared to the DVd group (78%). The clinical experts consulted indicated that the main reasons for treatment discontinuation (disease progression and AEs) were as expected, with ocular toxicity being the main concern associated with belantamab treatment. They also noted that several strategies have been developed to manage the ocular toxicity of belantamab and prevent permanent drug discontinuation. These strategies include the reduction of doses, the delay of treatments, the early detection of corneal toxicity, and the use of cooling masks. The BVd group had longer exposure to the study drug than the DVd group; therefore, differences in treatment discontinuation and exposure between the study groups may have influenced the interpretation of harms outcomes. In the DREAMM-7 trial, dose modifications were not permitted for daratumumab whereas dose modifications and delays were allowed for belantamab.

HRQoL was assessed only on the evaluable data with missing assessments handled implicitly based on a missing at random assumption. No strong conclusions could be drawn about the effect of BVd compared with DVd on HRQoL due to a high risk of attrition bias, particularly at longer follow-up. Including only evaluable patients in the model can introduce selection bias because this group may not be representative of the overall patient population. For example, patients who drop out or are excluded due to missing data or other factors may differ significantly from those who remain in the study, potentially skewing the results. The analysis method for HRQoL outcomes relied on the likely implausible assumption that data were missing at random, and no sensitivity analyses were conducted to assess robustness to different missingness mechanisms.

External Validity

According to the clinical experts consulted by CDA-AMC for this review, the patient population in the DREAMM-7 trial generally reflects patients in clinical practice in this setting. A total of 20.7% of the patients in the DREAMM-7 trial did not pass screening, primarily due to an inability to meet eligibility criteria. However, clinical experts consulted did not identify any major concerns regarding generalizability. Patients were eligible for inclusion in the DREAMM-7 trial if they met 1 of the measurability parameters, including having a serum M protein concentration of 0.5 g/dL or higher. The clinical experts noted that some patients with serum M protein concentration below 0.5 g/dL would still be eligible for the treatment with BVd in clinical practice. The clinical experts indicated that the exclusion criteria of the DREAMM-7 trial are consistent with clinical practice. However, the clinical experts viewed the exclusion of patients who were refractory to daratumumab as inappropriate because this does not align with clinical practice, where daratumumab is not commonly used as a comparator. They also noted that while daratumumab is a suboptimal comparator for belantamab, it was considered an appropriate comparator at the time the DREAMM-7 trial was designed. The clinical experts noted that bortezomib was administered twice a week in the trial (day 1, day 4, day 8, and day 11) whereas in clinical practice in Ontario, it is typically administered once a week for patient convenience. However, the clinical experts did not view this difference as a concern because efficacy and tolerability are generally comparable between the 2 dosing schedules.

According to the clinical experts consulted, the demographic and disease characteristics of the DREAMM-7 trial population were reflective of patients living in Canada with RRMM. The mean age of patients in the DREAMM-7 trial was 64 years, with clinical experts noting that in the real-world setting, transplant-eligible patients would likely be younger while patients with relapsed or refractory disease would typically be older. Most patients were white and not of Hispanic or Latino ethnicity; however, the clinical experts noted that this would not limit generalizability to patients in clinical practice in Canada. Most patients in the DREAMM-7 trial had an ECOG PS score of 0 or 1, and an R-ISS score of stage I and stage II. The clinical experts consulted indicated that this would be reflective of clinical practice in the second-line setting, but not in later relapse. As a result, the trial findings may be less generalizable to patients with more advanced disease or poorer performance status. They also pointed out that clinicians would prescribe belantamab to patients with an ECOG PS score of 2 and an R-ISS score of stage III. Although BVd is indicated for the treatment of patients with MM who have received at least 1 prior line of therapy, more patients in the trial had received 1 prior line of therapy (50.6%), and fewer had received 2 (23.7%) or 3 (14.2%) prior lines of therapy. However, the clinical experts consulted indicated that the evidence for PFS across subgroups still suggests a benefit for patients who have received 2 or more prior lines of therapy. All patients in the DREAMM-7 trial were previously exposed to lenalidomide or thalidomide, which is consistent with clinical practice. Only 34% of patients in the trial were refractory to lenalidomide. The clinical experts indicated that, in the real-world setting, the proportion of patients who are refractory to lenalidomide in second-line therapy and beyond would be higher because it is commonly used as maintenance therapy after achieving a response to initial treatment. They also noted that these patients may have a poorer response to second-line or third-line treatments compared to those who are not refractory to lenalidomide. According to clinical experts, in real-world settings, a proportion of relapsed patients are expected to be anti-CD38 refractory — a scenario that was not represented in the trial population. As a result, the DREAMM-7 trial population could overestimate treatment effectiveness and limit the generalizability of the trial results to a broader real-world population.

The clinical experts consulted by CDA-AMC for this review did not have any major concerns with the end points used in the trial. According to the clinical experts and patient and clinician group input, OS, PFS, clinical response outcomes, and HRQoL are the most important outcomes for assessing the response to treatment. Because the DREAMM-7 trial is ongoing, final OS data are not yet available. The difference in HRQoL outcomes between the groups was not formally assessed, which precludes definitive conclusions from being drawn about this important outcome. The analyses were performed in the HRQoL-evaluable population of patients who had evaluable assessment at baseline and follow-up time points, which limits the generalizability of the results.

In the DREAMM-7 trial, the study population was drawn from a number of sites around the globe, including Canada. The clinical experts indicated that there are no major concerns with generalizing the findings from the trial to the clinical setting in Canada.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group:^23,24

• “High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: We are moderately confident in the effect estimate — The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word ‘likely’ for evidence of moderate certainty (e.g., ‘X intervention likely results in Y outcome’).

• Low certainty: Our confidence in the effect estimate is limited — The true effect may be substantially different from the estimate of the effect. We use the word ‘may’ for evidence of low certainty (e.g., ‘X intervention may result in Y outcome’).

• Very low certainty: We have very little confidence in the effect estimate — The true effect is likely to be substantially different from the estimate of effect. We describe evidence of very low certainty as ‘very uncertain.’”

Following the GRADE approach, evidence from the RCT started as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. For this review, the target of the certainty of evidence assessment was based on the presence or absence of an important effect as informed by thresholds identified based on clinical expert input.

Results of GRADE Assessments

The GRADE assessments included an evaluation of the main outcomes considered important by clinicians, patient groups, and committee members.

Table 2 presents the GRADE summary of findings for BVd versus DVd in patients with MM.

Long-Term Extension Studies

No long-term extension studies were identified by the sponsor.

Indirect Evidence

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Objectives for the Summary of Indirect Evidence

The DREAMM-7⁷³ trial provides comparative evidence of the efficacy of BVd compared with DVd in patients with RRMM who had received at least 1 line of therapy. In the absence of direct comparative data for belantamab versus other relevant comparators used in clinical practice, aside from DVd, the feasibility of conducting ITCs through an NMA was assessed. ITCs were deemed feasible and performed for the primary outcome of PFS and other outcomes of interest, including OS and ORR.

Description of Indirect Comparison

One report that included ITCs was supplied by the sponsor. It included an SLR and an NMA, comparing BVd with cyclophosphamide plus carfilzomib plus dexamethasone, CyVd, DVd, EVd, hKd, hkDd, IhKd, Kd, PanoVd, PVd, SVd, and Vd.

Based on the prespecified eligibility criteria as outlined in Table 23, the sponsor conducted an SLR to identify studies investigating the efficacy of interventions used in the management of patients with RRMM who had received at least 1 prior therapy. The literature search was carried out in December 2021, with updates in May 2023 and January 2024. Multiple databases were searched to identify clinical trials. The study screening, study selection, and data extraction were conducted by 2 independent reviewers. Any discrepancies between the 2 reviewers were resolved by a third senior reviewer. Studies identified from the systematic literature search were first screened based on the title and abstract. Full-text screening was then carried out for studies selected from the title and abstract screening stage. Two independent reviewers extracted data and assessed the risk of bias in the included studies using the Cochrane risk-of-bias assessment tool for randomized trials.⁷⁴

The initial search strategy and the study selection criteria for the SLR were broad, incorporating comparators beyond those used in clinical practice in Canada to ensure relevance across multiple countries. At the full-text review stage, more restrictive selection criteria were applied. The list of studies identified in the SLR was narrowed down to include only studies with a connection to the DREAMM-7 study,⁷³ and further refined to include only treatments approved by the US FDA or the European Medicines Agency, as well as any regimens that are likely to serve as future comparators to the DREAMM-7 study’s regimen (BVd). The studies included patients with documented MM who have been previously treated with at least 1 line of therapy and had experienced disease progression during or after their most recent treatment. Only RCTs, including both primary and post hoc analyses, were considered. The intervention and comparator could involve any therapy or combination of treatments used for the treatment of RRMM.

Table 23: Study Selection Criteria and Methods for ITCs Submitted by the Sponsor

Anti-CD38 = anti–Cluster of Differentiation 38; ADC = antibody-drug conjugate; AE = adverse event; ASCT = autologous stem cell transplant; cilta-cel = ciltacabtagene autoleucel; CLR-131 = iopofosine I-131; CR = complete response; CRS = cytokine release syndrome; DoR = duration of response; GSK = GlaxoSmithKline; GSK’916 = GlaxoSmithKline 916; HDAC = histone deacetylase; HLH/MAS = hemophagocytic lymphohistiocytosis/macrophage activation syndrome; ICANS = immune cell-associated neurotoxicity syndrome; ide-cel = idecabtagene vicleucel; INAHTA = International Network of Agencies for Health Technology Assessment; ITC = indirect treatment comparison; lete-cel = letetresgene autoleucel; MM = multiple myeloma; MR = minimal response; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; PD = progressive disease; PFS = progression-free survival; PFS2 = progression-free survival-2; PI = proteasome inhibitor; PR = partial response; RCT = randomized controlled trial; RoB 2 = risk of bias 2; RRMM = relapsed or refractory multiple myeloma; SAE = serious adverse event; sCR = stringent complete response; TRAEs = treatment-related adverse events; TTBR = time to best response; TTNT = time to next therapy; TTP = time to progression; TTSR = time to systemic response; TTTF = time to treatment failure; URTI = upper respiratory tract infection; VGPR = very good partial response.

Sources: Sponsor-submitted indirect treatment comparison report.⁷⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

ITC Design

The primary objective of the ITC was to assess the comparative efficacy of BVd in a population matching the ITT population of the DREAMM-7 trial, relative to other comparators for which PFS data were available. The secondary objectives of the NMA were to assess the comparative efficacy of BVd in a population of prior lenalidomide-exposed patients, based on available data for OS and ORR, and in a population of lenalidomide-refractory patients, relative to comparators for which data were available for PFS and ORR.

The following outcomes were reported to address the objectives of the NMA: PFS, OS, and ORR for the ITT population, and PFS and ORR for the lenalidomide-exposed and lenalidomide-refractory populations. PFS was defined using IMWG guidelines²⁵ as the time from randomization to the first documented disease progression or death due to any cause, whichever occurs first. OS was defined as the time from randomization to death due to any cause. ORR was defined as the percentage of patients with a confirmed PR or better (i.e., VGPR, CR, or sCR), in line with the IMWG response criteria²⁵ and recorded by an IRC. To meet submission timelines, the number of outcomes was narrowed, excluding MRD negativity, CR or better, VGPR or better, and progression-free survival-2. OS was not analyzed for the lenalidomide-exposed or lenalidomide-refractory populations due to the absence of reported data in the CASTOR^77,78 and LEPUS⁷⁹ studies, which prevented the formation of a connected network.

ITC Analysis Methods

Bayesian NMAs were performed using random-effects or fixed-effects models for all analyses. A fixed-effects model was applied for the base-case analysis, a random-effects model was used for an exploratory analysis. For time-to-event end points, it was assumed that the treatment effect (log-HRs) follows a normal distribution. For binary end points, a standard binomial model with a logit link was used. The primary analysis was performed for PFS, OS, and ORR to assess the efficacy of BVd relative to comparators in the ITT population. The secondary analyses were performed for PFS and ORR to assess the efficacy of BVd relative to comparators in the lenalidomide-exposed and lenalidomide-refractory populations. Based on potential biases identified in the feasibility assessment, subgroup analyses were performed for PFS in patients with 1 prior line of therapy and high cytogenetic risk. The prior distributions for between-study heterogeneity were derived from the study by Turner et al.⁸⁰ and were used as informative priors to directly inform the random-effects NMA. Convergence to the target posterior distributions was assessed using the Gelman-Rubin statistic.⁸¹ Convergence was monitored quantitatively using the latest implementation Gelman-Rubin diagnostic (Rhat) based on 4 chains. For each analysis, the goodness of fit to the model was assessed by using the deviance information criterion and residual deviance. The presence of clinical and methodological heterogeneity was assessed through a feasibility assessment in which the population and study characteristics were compared. The inconsistency assessment for the NMA evaluated the consistency between estimates from direct and indirect evidence. League tables were used to present the results of the pairwise comparisons from the NMA between treatments. The multi-NMA package version 0.7.0 and R version 4.2.1 or above via Rstudio were used to perform the analysis.

Table 24: ITC Analysis Methods

BVd = belantamab mafodotin plus bortezomib and dexamethasone; CR = complete response; CrI = credible interval; DIC = deviance information criterion; HR = hazard ratio; IA = investigator assessed; IMWG = International Myeloma Working Group; IRC = independent review committee; ITC = indirect treatment comparison; log-HR = log–hazard ratio; NMA = network meta-analysis; OR = odds ratio; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; RRMM = relapsed or refractory multiple myeloma; sCR = stringent complete response; TEM = treatment effect modifier; VGPR = very good partial response; vs. = versus.

Note: Several sensitivity and subgroup analyses were planned but deprioritized due to time constraints.

Sources: Sponsor-submitted indirect treatment comparison report.⁷⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results of ITC

Summary of Included Studies

From the original and update searches, conducted in December 2021 and February 2024, respectively, 63 unique studies were identified from a total of 204 publications. No geographic regions were specified when conducting the SLR.

The feasibility assessment was originally conducted using the December 2021 SLR results. The evidence network was then reduced to include only studies with a connection to the DREAMM-7 treatment groups.²⁶ Of the 47 studies identified, 16 studies investigating 17 treatments were found to form a connected network, anchored through 4 common treatments: Vd, hKd, Kd, and DVd. The network was further restricted to include only treatments that are approved by the US FDA or the European Medicines Agency and any regimens likely to be a future health technology assessment comparator to BVd, the DREAMM-7²⁶ study regimen. Based on these criteria, 4 of the 17 treatments (from 6 studies) did not meet the inclusion criteria and were excluded from the analysis: venetoclax plus bortezomib plus dexamethasone; thalidomide plus dexamethasone; and tabalumab (100 mg and 300 mg) plus dexamethasone plus bortezomib.^82-87 The restricted evidence network based on 13 studies, including the ARROW,⁸⁸ BOSTON,⁸⁹ CANDOR,^52,90 CASTOR,^77,78 ENDEAVOR,^91,92 IKEMA,²⁹ OPTIMISMM,⁵⁰ GEM-KyCyDex,⁹³ LEPUS,⁷⁹ PANORAMA-1,^94,95 NCT01478048,⁹⁶ NCT00813150,⁹⁷ and DREAMM-7²⁶ studies, is presented in Figure 9 and Table 28. The SLR updates identified new studies, all of which were deemed infeasible to include in the network based on an initial feasibility assessment. The majority of new studies were either single-arm trials, lacked a connection to the network, or did not report any outcomes of interest.

Lenalidomide plus dexamethasone, daratumumab plus lenalidomide and dexamethasone, and carfilzomib plus lenalidomide and dexamethasone are also available treatment options for patients with RRMM in clinical practice in Canada. However, because the majority of patients in Canada receive a lenalidomide-based therapy in the first line of therapy, these treatments were not considered relevant comparators to BVd in the second-line settings and beyond.²⁸

Figure 9: Network of Evidence Based on Studies Identified in the SLR

* mg/m²

** mg/kg

B-Vd = belantamab mafodotin plus bortezomib and dexamethasone; Be = belantamab; Cy = cyclophosphamide; CyKd = cyclophosphamide plus carfilzomib and dexamethasone; CyVd = cyclophosphamide plus bortezomib and dexamethasone; d = dexamethasone; D = daratumumab; DVd = daratumumab plus bortezomib and dexamethasone; E = elotuzumab; EVd = elotuzumab plus bortezomib and dexamethasone; hK = high-dose carfilzomib; hKd = high-dose carfilzomib plus dexamethasone; hKDd = high-dose carfilzomib plus daratumumab and dexamethasone; IhKd = isatuximab plus high-dose carfilzomib and dexamethasone; K = carfilzomib; Kd = carfilzomib plus dexamethasone; P = pomalidomide; Pano = panobinostat; PanoVd = panobinostat plus bortezomib and dexamethasone; PVd = pomalidomide plus bortezomib plus dexamethasone; S = selinexor; SLR = systematic literature review; SVd = selinexor plus bortezomib plus dexamethasone; V = bortezomib; Vd = bortezomib plus dexamethasone.

Note: The Vd treatment group also included placebo.

The carfilzomib and dexamethasone dosage and treatment pattern are identical in each GEM-KyCyDex study group and in the low-dose treatment group for the ARROW trial.

Sources: Sponsor-submitted indirect treatment comparison report.⁷⁶ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Feasibility Assessment

A feasibility assessment was conducted to inform on whether an NMA between the RCTs identified in the SLR would be feasible. This included determining whether the treatment effect modifiers (TEMs) — variables that can influence the effect of treatment on outcomes — along with study characteristics, population characteristics, and RRMM-specific characteristics, were homogenous.

The potential TEMs in patients with MM were identified through a targeted literature review and were based on the opinion of the sponsor’s clinical expert. TEMs in RRMM have been studied in a Bayesian NMA by Dimopoulos et al. (2018)⁹⁸ and an NMA and simulation study by Rose et al. (2023).⁹⁹ Dimopoulos et al. found more than 1 previous line of therapy to be a relevant TEM while Rose et al. found weak evidence of an interaction for within-trial TEMs for PFS by line of therapy and prior use of IMiD or lenalidomide-refractory status.⁹⁹ The potential list of TEMs included the prior line of therapy, the refractory status to the specific drug in the trial, the ISS stage, the cytogenetic risk profile, extramedullary disease, creatine clearance, the time from diagnosis, age, gender, ethnicity, comorbidities, and ECOG PS score.

To identify which variables are TEMs in the RRMM population, efficacy results by subgroup for potential TEMs and prognostic variables were explored for each study. PFS was used as the primary way to evaluate whether a factor was a TEM because it was the primary end point of the DREAMM-7 study. The TEMs explored for inclusion were those variables that exhibited imbalances between studies in the ITT populations and for which comparisons were feasible. The variables explored included:

• prior therapy

• prior IMiD exposure

• ISS stage

• ECOG PS

• prior lenalidomide exposure

• prior daratumumab use

• cytogenic risk profile.

None of the studies in the network stratified results by prior daratumumab exposure. Consequently, this variable was not available for exploration as a TEM. Upon assessing the available PFS data by subgroup across the studies included in the NMA, the results indicated that prior therapy, ECOG PS score, and ISS stage may be TEMs in the RRMM population.

Study Characteristics

All studies were phase III RCTs, except for 2 studies, which were phase II RCTs (Table 28). The trials were published between 2014 and 2024, during which time the treatment landscape had evolved. Crossover was permitted in 2 trials (the BOSTON⁸⁹ and LEPUS studies⁷⁹), which poses a risk of bias for OS results. Substantial variation was observed in the follow-up time at which outcomes were evaluated in the studies, with follow-up time for reporting PFS ranging from 5.6 months (the PANORAMA-1 trial^94,95) to 47 months (the CASTOR trial^77,78). In the case of the OPTIMISMM study,⁵⁰ 58.3% of patients in the Vd group subsequently received PVd, and therefore a crossover-adjusted OS HR was used to minimize the potential bias. The PANORAMA-1 study^94,95 was the only study of the network to include a placebo as part of a treatment regimen. Most treatment nodes in the network were only connected by 1 study. In the DREAMM-7 trial,²⁶ PFS was recorded by both IRC (primary end point) and IA, whereas the remaining 12 studies used an IRC (8 studies) or IA (4 studies) for recording this outcome.

The dosing of dexamethasone across the network exhibited some differences in either dose or cycle structure of dexamethasone between the ARROW,⁸⁸ CASTOR,^77,78 CANDOR,^52,90 and DREAMM-7 studies²⁶ (Table 28, Appendix 1). In particular, the following differences emerged:

• The dexamethasone dose was consistent for most of the network at 20 mg, except in the ARROW⁸⁸ and CANDOR studies,^52,90 in which patients received 40 mg doses. However, in the CANDOR study,^52,90 there was an option to reduce the dosage of dexamethasone to 20 mg for patients aged 75 years or older.

• The dexamethasone dosing structure was equivalent for the CASTOR study^77,78 and both groups of the DREAMM-7 study²⁶ with doses on day 1, day 2, day 4, day 5, day 8, day 9, day 11, and day 12. The GEM-KyCyDex⁹³ study had a slightly different dosing structure of dexamethasone with doses on day 1, day 2, day 8, day 9, day 15, and day 16 for both treatment groups^77,78 and both groups of the DREAMM-7 study²⁶ with doses on day 1, day 2, day 4, day 5, day 8, day 9, day 11, and day 12. The GEM-KyCyDex⁹³ study had a slightly different dosing structure of dexamethasone with doses on day 1, day 2, day 8, day 9, day 15, and day 16 for both treatment groups.

• Dosing frequency was comparable across the network (once weekly and twice weekly).

Population Characteristics

Table 25 provides a comparison of the characteristics of the populations included in studies comprising the network. The median ages of patients and the proportion of males were comparable across the studies. Race data were only available for only 6 of the 13 studies; comparability could not be fully assessed, but race distribution was comparable across those 3 studies. The time since diagnosis was available for 7 of the 13 studies; comparability could not be fully assessed but was consistent across studies that reported this characteristic.

Table 25: Similarity of Patient Population Characteristics