CADTH Reimbursement Review

Idecabtagene Vicleucel (Abecma)

Sponsor: Celgene Inc., a Bristol Myers Squibb Company

Therapeutic area: Multiple myeloma

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Ethics Review

Clinical Review

Executive Summary

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

Introduction

Multiple myeloma (MM) is a malignant, clonal bone marrow plasma cell disorder that causes clinical features of renal failure, bone lesions, hypercalcemia, and bone marrow suppression. It results from excessive production of monoclonal proteins as well as direct tumour cell effects.^1,2 Multiple myeloma is the second most common hematologic malignancy and represents approximately 1% of all cancers and roughly 10% to 15% of all hematologic malignancies. It primarily affects older individuals and is extremely rare in patients younger than 30 years old.^3,4 In Canada, the median age at diagnosis is 70 years, with an average national incidence rate of 5.42 cases per 100,000 individuals per year, a steady rise in the annual rate of 0.96 cases per million individuals per year, and an estimated 3,400 new cases annually.^1,5 At the provincial level, Quebec and Ontario have the highest MM incidence rates in Canada (5.82 cases per 100,000 individuals per year and 5.66 cases per 100,000 individuals per year, respectively). The mean national mortality rate in Canada stands at 3.57 deaths per 100,000 individuals per year.¹ Patients are classified with relapsed or refractory multiple myeloma (RRMM) when there is a recurrence of disease after prior response on the basis of objective laboratory and radiological criteria.⁶ From there, patients begin moving to successive therapies as they advance in the course of the disease with further relapses and more lines of therapies.

The goal of treatment in patients with RRMM is to achieve disease control with acceptable toxicity and patient-defined health-related quality of life (HRQoL). Meeting this goal relies on effective systemic chemotherapy and supportive measures such as pain control, antibiotics, kyphoplasty, radiation therapy, dialysis, and psychosocial support. Although emerging options are increasingly available,⁷ no standard of care has been fully established and, despite advances, choosing an optimal strategy at relapse remains a challenge for clinicians. Currently, the options for patients with RRMM are second to fifth lines of therapies, which include several drugs. The sequencing of treatments depends largely on the type of response to previous therapies. Regimens commonly used in the Canadian landscape include combinations of daratumumab plus bortezomib plus dexamethasone, bortezomib plus dexamethasone (Vd), pomalidomide plus dexamethasone, and carfilzomib plus dexamethasone (Cd), each of which is used when there is resistance to lenalidomide. Regimens used when there is resistance to bortezomib-based therapies include combinations of daratumumab, lenalidomide, and dexamethasone; carfilzomib, lenalidomide, and dexamethasone; and lenalidomide and dexamethasone.^7-9

Idecabtagene vicleucel is a cell therapy formed by chimeric antigen receptor (CAR)-positive T cells directed against the B-cell maturation antigen (BCMA). The indication is for the treatment of adult patients with MM who have received at least 3 prior therapies, including an immunomodulatory drug, a proteasome inhibitor, and an anti-CD38 antibody, and who are refractory to their last treatment. Idecabtagene vicleucel is provided as a single-dose, 1-time treatment in 1 or more patient-specific infusion bags containing a suspension of 275 × 10⁶ to 520 × 10⁶ CAR T cells. The target dose is 450 × 10⁶ CAR T cells within a range of 275 × 10⁶ to 520 × 10⁶ CAR T cells. Pretreatment includes the use of lymphodepleting chemotherapy (LDC) with cyclophosphamide 300 mg/m² IV and fludarabine 30 mg/m² IV for 3 days. Idecabtagene vicleucel infusion is to be administered 2 days after completion of LDC. The objective of this report is to perform a systematic review of the beneficial and harmful effects of idecabtagene vicleucel for the treatment of adult patients with MM who have received at least 3 prior therapies, including an immunomodulatory drug, a proteasome inhibitor, and an anti-CD38 antibody.

Stakeholder Perspectives

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

Patient Input

Myeloma Canada submitted the patient input for this review. Founded in 2005, Myeloma Canada is the only national charitable organization created by and for Canadians affected by MM. The organization works to improve the lives of those affected by myeloma. Information from this input was gathered through a patient survey. The survey was accessed through email and social media from December 17, 2020, to January 4, 2021. A total of 388 individuals with myeloma responded to the survey.

Overall, patients described how myeloma, as part of their lives, affects major decisions for themselves and their family members. Patients who have been heavily treated expect new treatment options to provide prolonged remission, not having to be on their current treatment because they are in remission, better quality of life, better overall health, better control of myeloma symptoms, and fewer side effects. More than half of patients responding to the survey set optimistic expectations for new options such as CAR T-cell therapies. Among 6 patients with previous experience with a CAR T-cell therapy, 5 considered the intervention effective, and 3 stated that the burden of receiving it was tolerable, with decreased cognitive abilities and neutropenia as the most commonly described effects. “I’m still alive,” “so far in very, very good remission,” and “I would have otherwise run out of options” were some of the phrases obtained from the survey, along with some less-encouraging responses, such as “didn’t get the depth of response as we were hoping.”

The patient group highlighted that many Canadians are looking for new options for effective treatments, particularly when they reach the point of multiple treatments with remission and refractoriness to the available regimens. Patients were willing to trade some of the burdens of the new therapy as they assigned more value to outcomes such as prolonged survival and quality of life (QoL).

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical experts consulted by CADTH agreed that not all patients with MM respond to available treatments; most eventually become refractory to all treatment options. These patients have a short life expectancy, and their remaining life expectancy is often highly medicalized. Treatments are needed to control the disease and provide patients with a break from therapy. There are few options (other than palliative or supportive care) after the third and fourth line of therapy, leaving a major unmet need. The majority of CAR T-cell treatments are financially burdensome both at the level of the individual patient and the public budget.

In the current treatment paradigm, CAR T-cell therapies would be well positioned in the third line (current Canadian approval) and fourth line (older algorithms). The clinical experts noted that, when daratumumab, bortezomib, lenalidomide, and dexamethasone are approved as front-line treatments, CAR T cells could become a second-line option. The experts agreed that patients best suited for idecabtagene vicleucel treatment would be identified by following the inclusion and exclusion criteria from the study, together with an individual clinical assessment and judgment based on prior therapies, patient comorbidities, and fitness. Clear, transparent, and evidence-based recommendations will be important for decision-making. All standard diagnostic and baseline tests are available broadly in Canada.

The eligible population will be small because many patients already reach a palliation stage by the fourth line of treatment due to rapid disease progression, age, frailty, or toxicities from prior treatments. The eligible patients (i.e., those with a good performance status and those who have been previously treated with or are refractory to proteasome inhibitors, immunomodulatory imide drugs (IMIDs), and CD38 monoclonal antibodies) appear to be the most appropriate group with the greatest unmet need. Clinical experts stated that the patients best suited for this treatment would be those with an Eastern Cooperative Oncology Group (ECOG) status of 0 or 1, good organ status, and a willingness to travel (this therapy will not be offered in smaller centres). Those with relapsed or refractory disease should be prioritized first. Younger, fit patients with the least number of prior therapies (3 prior lines) are the most likely to benefit.

According to the experts consulted by CADTH, the patients least suitable for the treatment under review are those with a poor performance status, and those with other options available with a reasonable duration of disease control. Additionally, patients who are frail, elderly, have significant renal failure, neurologic disease, and an ECOG status of 2 or greater related to myeloma would not be suitable candidates for idecabtagene vicleucel treatment. The experts suggested that progression-free survival (PFS) longer than 6 months or overall survival (OS) for more than 1 year without life-threatening toxicities would be considered a clinically meaningful response to treatment.

The clinical experts consulted by CADTH agreed that idecabtagene vicleucel must be administered in a setting supervised by specialists trained in bone marrow transplant, a tertiary centre that is well-versed in complications related to stem cell transplant, or a centre that can accommodate phase I trials, with adequate intensive care, emergency, neurology, and infectious disease departments. They also agreed that a hematologist or an oncologist with specific experience in cellular therapy is required to diagnose, treat, and monitor patients who might receive idecabtagene vicleucel.

Clinician Group Input

Two clinician groups provided information for the reimbursement process: the Canadian Myeloma Research Group (CMRG), a charitable organization supported by myeloma physicians from 22 academic centres in Canada, and the Ontario Health (previously Cancer Care Ontario) Hematology Cancer Drug Advisory Committee (DAC), which provides evidence-based clinical and health system guidance on drug-related issues to provincial drug reimbursement programs and others.

Overall, both groups agreed with the clinical experts consulted by CADTH with respect to unmet needs in those whose life expectancy is limited (i.e., triple-class exposed patients) and both noted that only a few options are available for heavily treated patients with RRMM. The groups added that there is no sequencing information available beyond that used in the triple-class exposed population.

The clinician groups noted that, although no data are available to indicate which patients are most likely to exhibit a response, patients with good performance status, organ function, minimal or no comorbidities, robust blood counts, low tumour burden, and indolent disease are most likely to experience the best outcomes.

The clinician groups noted that the drug under review will be delivered at tertiary hospitals or transplant centres with expertise in cellular therapy. They added that idecabtagene vicleucel patients would require an inpatient bed for approximately 2 weeks (or longer if complications occur) and may require readmission.

Drug Program Input

The Provincial Advisory Group (PAG) provided questions considered important for decision-making.

• Which patient (or disease) characteristics would be important in determining who should be treated with this therapy?

• If capacity limitations exist, how to prioritize which patients should be offered idecabtagene vicleucel?

• Do patients need to have disease considered refractory to an IMiD, proteasome inhibitor, and an anti-CD38 antibody to receive idecabtagene vicleucel?

• It is important to discuss whether or not there should be re-treatment (provided data are scarce). The sponsor suggests idecabtagene vicleucel will be a single-dose, 1-time treatment, but re-treatment was allowed in the KarMMa trial. This would imply additional costs (i.e., due to cryopreservation, cell processing, tocilizumab, intensive care unit and hospital stays, etc.).

The clinical experts consulted by CADTH stated the characteristics of patients with MM who should receive idecabtagene vicleucel as above, to those with prior treatment with an IMiD, a proteasome inhibitor, or an anti-CD38 antibody. The only unresolved concern was the exact line of therapy to recommend when initiating CAR T-cell treatment. The clinical experts consulted by CADTH anticipated that prioritization of the therapy would rarely be a concern, in which case deliberations should be made on a case-by-case basis; the same conclusion was drawn for the re-treatment considerations.

PAG identified a possible issue with the prescription of idecabtagene vicleucel, specifically that access would be limited to jurisdictional capacity. Given that the sponsor plans to have 12 sites across Canada, out-of-province care may be needed for proper administration of the drug under review. Also, staff will need to be trained and specialized centres accredited by the sponsor, limiting the availability of clinics where the therapy can be provided.

PAG noted that there may be potential for “indication creep” to earlier lines of therapy. For example, triple-class exposure may occur in the first 2 lines of therapy, although the sponsor’s proposed indication does not establish a specific line of therapy. The clinical experts consulted by CADTH agreed that it would be important to establish the appropriate line of therapy for idecabtagene vicleucel (currently their recommendation was third or fourth) and which patient populations would be outside the scope of the reviewed indication. Furthermore, idecabtagene vicleucel may change the place in therapy of other (comparator) drugs and in subsequent lines as more data become available.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

The KarMMa study (N = 140) was a phase II, open-label, multi-centre, single-arm trial conducted at 20 sites across Europe and North America, including 1 in Canada. Eligible patients were adults who had received at least 3 previous regimens for MM, including an immunomodulatory drug, a proteasome inhibitor, and an anti-CD38 antibody, and who had disease that was refractory to their last regimen (progression within 60 days after the last dose) according to International Myeloma Working Group (IMWG) criteria. The main objective of the study was to evaluate the efficacy, defined as the overall response rate (ORR), of idecabtagene vicleucel in patients with RRMM. Secondary objectives included the assessment of the safety of idecabtagene vicleucel and other measures of efficacy, such as the complete response rate (CRR), duration of response (DoR), time to response (TTR), time to progression (TTP), minimal residual disease (MRD)-negative status, OS, PFS, and HRQoL. Harms evaluated included adverse events (AEs), health care utilization, and others of special interest for this review, such as cytokine release syndrome (CRS), febrile neutropenia, and neurotoxicity.

All 140 enrolled patients underwent leukapheresis (i.e., drug manufacturing). If necessary, a bridging therapy (with corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies alone or in combination) was provided while the idecabtagene vicleucel was manufactured but had to be stopped at least 14 days before LDC. All patients were required to undergo LDC with cyclophosphamide and fludarabine 5 days before infusion with idecabtagene vicleucel. Of the 140 patients enrolled, 128 received idecabtagene vicleucel infusion and were included as the idecabtagene vicleucel–treated population for the primary analysis of efficacy and safety. Patients received target doses of 150 × 10⁶, 300 × 10⁶, or 450 × 10⁶ CAR T cells and were followed for at least 24 months and then asked to participate in a separate long-term follow-up study. The median age of the 140 enrolled patients was 60.5 years (range = 33 to 78), 82 (58.6%) were male, the median years since diagnosis was 6 years (range = 1 to 18), 46 (32.9%) were at high cytogenetic risk, and 131 (93.6%) had received a stem cell transplant.

Efficacy Results

The ORR, CRR, DoR, OS, and HRQoL were considered the most relevant outcomes by the clinical experts and clinician groups. In addition, patient groups considered improved HRQoL (e.g., fewer symptoms, higher functionality, and fewer side effects from treatments) to be highly valued outcomes.

At a median follow-up of 11.3 months, the KarMMa trial met its primary end point, achieving an ORR above the null hypothesis of 50% established in the protocol. The ORR in the population treated with idecabtagene vicleucel was 73.4% (95% confidence interval [CI], 65.8% to 81.1%; P < 0.0001) and 81.5% (95% CI, 68.6% to 90.7%) in the group with a target dose of 450 × 10⁶ CAR T cells, respectively. Given that the test for ORR (the primary end point) was positive, the key secondary efficacy end point of CRR was tested against the null hypothesis of 10% or less — as prespecified — rejecting the null hypothesis, with 40 of 128 patients achieving a complete response equivalent to a CRR of 31.3% (95% CI, 23.2% to 39.3%; P < 0.0001), while in the group with a target dose of 450 × 10⁶ CAR T cells, the CRR was 35.2% (95% CI, 22.7% to 49.4%). These values from the primary and key secondary end points were considered clinically meaningful by the clinical experts.

The DoR, TTR, and TTP were evaluated in the KarMMa trial as secondary end points. In the treated population, the median DoR was 10.6 months (95% CI, 9.0 to 11.3), the TTR was 1 month (range = 0.5 to 8.8), and the TTP was 8.9 months (95% CI, 6.0 to 11.6), while in the group with a target dose of 450 × 10⁶ CAR T cells, the median DoR was 11.30 months (95% CI, 9.17 to 11.43) and the TTR was 1.0 months (range = 0.9 to 2.0); TTP was not assessed in the 450 × 10⁶ CAR T cells group. Likewise, an MRD-negative status with a CR or better was observed in 31 of 128 patients (24.2%; 95% CI, 17.1% to 32.6%) treated with idecabtagene vicleucel, and in 13 of 54 of patients in the group with a target dose of 450 × 10⁶ CAR T cells. Although these results may suggest a beneficial effect in patients treated with idecabtagene vicleucel and were considered likely to constitute a meaningful impact from the perspective of the clinical experts, none of these secondary end points were adjusted for multiplicity and should be interpreted with caution.

Both experts and patients considered OS an important and valued outcome and it is commonly used in clinical trials conducted in patients with MM. In the treated population, the observed median OS was 18.2 months (95% CI, 18.0 to not estimable [NE]), with 76% of individuals alive at 12 months; the clinical experts considered this a beneficial effect estimate for patients with the characteristics included in the KarMMa trial (i.e., those with a median of 6 years since diagnosis, with relapses, and who have been heavily treated). The OS was NE in patients in the group with a target dose of 450 × 10⁶ CAR T cells. Similarly, based on the Kaplan–Meier (KM) estimates, the median PFS was 8.6 months (95% CI, 5.6 to 11.3), with 34% of patients event-free at 12 months in the idecabtagene vicleucel–treated group, and 11.3 months (95% CI, 8.8 to 12.4) in the group with a target dose of 450 × 10⁶ CAR T cells, denoting a likely meaningful benefit for patients treated with idecabtagene vicleucel.

Outcomes involving HRQoL were frequently mentioned in the input submitted by patient groups and by clinician groups. Patients put a high value on decreasing symptoms, better physical functioning, less pain, and fewer side effects. Results of the KarMMa trial suggest that idecabtagene vicleucel treatment may be associated with improvements in the fatigue, pain, physical functioning, and global health/QoL subscales of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) by reaching points of meaningful clinical significance above the thresholds of probable benefit according to minimal important differences (MIDs) established in the literature. On average, no clinically meaningful deterioration in the EORTC QLQ-C30 cognitive functioning and the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Myeloma Module with 20 items (EORTC QLQ-MY20) disease symptoms and side effects subscales were observed posttreatment. When addressing the EuroQol 5-Dimensions 5-Levels (EQ-5D-5L) subscales, no changes from baseline were observed. These findings appear to suggest that idecabtagene vicleucel treatment may be associated with improvements or stabilization in most primary HRQoL metrics. However, these results were observed over the first and middle part of the study follow-up, eventually decreasing to baseline levels, with added uncertainty due to imprecision (i.e., fewer patients at the end of the study) and the fact that all HRQoL outcomes in the KarMMa study were secondary end points assessed without adjustment for multiplicity.

Sensitivity analyses were based on whether the FDA or European Medicines Agency (EMA) rules for censoring DoR, TTP, and PFS outcomes were used; and on comparisons using an independent response committee (IRC) versus investigator adjudication of end points (for ORR, CRR, and TTP). Finally, scenarios were presented when analyzing the enrolled population against the population treated with idecabtagene vicleucel (for DoR, CRR, PFS, and OS). No notable differences from the primary effect estimates were observed for any of these outcomes using any of the sensitivity analyses.

Except for the ORR and CRR outcomes, no other secondary end points were considered in the adjustment for multiplicity and should be interpreted with caution.

In an updated data submission (cut-off date of January 14, 2020) with a median follow-up of 13.3 months,¹⁰ no significant changes were noted in the ORR, CRR, DoR, TTR, MRD, OS, or PFS. These updated results, described in Appendix 3, were not adjusted for alpha errors nor were they used for hypothesis testing.

Harms Results

AEs were reported in all 128 (100%) patients treated with idecabtagene vicleucel with. Most AEs, with the exception of hypogammaglobulinemia and infections, occurred within 8 weeks after infusion. The most commonly reported AEs were hematologic toxic effects, including neutropenia in 117 patients (91.4%), CRS in 107 patients (83.6%), anemia in 89 patients (69.5%), and thrombocytopenia in 81 patients (63.3%). A total of 86 patients (67.2%) had at least 1 serious adverse event (SAE). The most frequently reported SAEs (≥ 5% of patients) were CRS in 22 patients (17.2%), general physical health deterioration in 13 patients (10.2%), pneumonia in 11 patients (8.6%), and febrile neutropenia in 9 patients (7.0%).

Eight patients died after leukapheresis and before receiving idecabtagene vicleucel infusion: 5 patients (3.6%) died after leukapheresis and before starting LDC, while 3 patients (2.1%) died after starting LDC and before receiving idecabtagene vicleucel infusion. In the idecabtagene vicleucel–treated population as of the data cut-off date, 34 patients (26.6%) died on or after idecabtagene vicleucel infusion, with 24 of those death which were attributed to the malignant disease or complications due to the malignant disease under study.

Harms of special interest were identified according to the protocol for this review. Febrile neutropenia was present in 21 patients (16.4%) of the idecabtagene vicleucel–treated population. A total of 23 patients (18.0%) in the idecabtagene vicleucel–treated population had investigator-identified neurotoxicity on or after idecabtagene vicleucel infusion. Evidence of CRS on or after the idecabtagene vicleucel infusion was present in 107 patients (83.6%).

In an updated cut-off data submission (cut-off date: January 14, 2020),¹⁰ no differences in the number of patients with AEs were noted. Ten more deaths (44 [34.4%]) were reported at this cut-off date and 23 more cases of febrile neutropenia (for a total of 30 of 128 treated patients [16.3%]) were reported. The other harms of special interest were not significantly different at this cut-off date.

Critical Appraisal

This single-arm, phase II study lacks a comparator group, which limits confidence in the effect estimates due to an increase in the risk of bias due to confounding fluctuations in health status, and unidentified prognostic factors that could affect subjectively assessed outcomes. The risk of attrition bias is low in the primary and key secondary end points, given that all patients in the population treated with idecabtagene vicleucel were included in the analysis. Outcomes regarding HRQoL were characterized by large amounts of missing data at baseline, with a significant proportion missing at a later time. Further, the open-label nature of the study design may introduce bias for patient-reported outcomes, such as HRQoL measures.

The only outcomes adjusted for multiplicity were the ORR and CR, and results of other secondary outcomes (DoR, TTR, TTP, MRD status, OS, PFS, and HRQoL) must therefore be interpreted with consideration for the risk of type I error.

It is unknown if bridging therapy interventions could affect the outcomes evaluated in these patients after they received idecabtagene vicleucel treatment.

Subgroups were planned a priori, but no formal subgroup effect tests were conducted due to the small number of patients and events (imprecision) expected.

The clinical variables and patient baseline characteristics from the KarMMa trial were generally representative of Canadian patients with RRMM who are refractory to IMiDs, proteasome inhibitors, and anti-CD38 drugs. The administration procedure of idecabtagene vicleucel will require additional actions (i.e., manufacturing, bridging therapies, and LDC); experts considered the administration procedure followed in the KarMMa study to be as anticipated in Canada.

Indirect Comparisons

Description of Studies

Two sponsor-submitted analyses are included that compare the information from the single-arm KarMMa study to observational evidence obtained from individual patient data or aggregated published data.

The first analysis (NDS-MM-003)¹² is a comparison of the effects of idecabtagene vicleucel in the single-arm KarMMa study (MM-001) against other currently administered treatments for RRMM obtained from a set of real-world (RW) patient-level data collected from various sources, including databases and clinical sites from the US and Europe. To decrease the imbalances or differences in patients from the real-world evidence (RWE) when compared with the KarMMa population, propensity scores were created.

From the RWE databases, eligibility criteria were used to collect a first cohort of RRMM patients (eligible cohort) and a final cohort (matched cohort) of RRMM patients with the intention to use an estimated 2:1 matching of RW patients with RRMM to be compared with patients receiving idecabtagene vicleucel in the single-arm KarMMa study. Due to the strict criteria, fewer patients were available for the matched RRMM cohort, requiring a change in the statistical analysis. The investigators decided to use the eligible RRMM cohort for the main analysis, using a trimmed stabilized inverse probability treatment weighting (IPTW) method, leaving the matched-paired and untrimmed IPTW as supportive analyses.

The outcomes evaluated were the ORR, CRR, very good partial response (VGPR) rate, PFS, and OS. Baseline characteristics of the idecabtagene vicleucel arm and the RWE eligible RRMM cohort indicated an older population in the eligible RRMM cohort, with a higher risk of cytogenetic abnormalities, and fewer previous antimyeloma therapies. Prior to the IPTW adjustment, with the exception of age, the imbalance of the prognostic factors and prior treatments suggested the eligible RRMM cohort was a less heavily pretreated and less refractory population with a more favourable prognosis overall compared with the idecabtagene vicleucel cohort.

The second analysis is a sponsor-submitted matching-adjusted indirect comparison (MAIC)¹³ that compares the idecabtagene vicleucel–treated population from the KarMMa trial to aggregated data from a published RWE study (MAMMOTH). These analysis aim to provide a comparator arm, with adjustments for differences in baseline characteristics, prognostic factors, and effect modifiers, and produce an adjusted effect estimate for decision-makers and stakeholders, given the lack of a direct, head-to-head comparison of idecabtagene vicleucel to relevant comparators.

Efficacy Results

The NDS-MM-003 analysis included 128 patients from the original KarMMa trial (idecabtagene vicleucel–treated population) and compared them against 190 patients in the RWE databases (the eligible RRMM cohort) with similar eligibility criteria. The obtained ORR was lower in the eligible RRMM cohort compared with the idecabtagene vicleucel cohort (32.2% versus 76.4%; risk ratio (RR) = 2.4; 95% CI, 1.7 to 3.3; P < 0.0001).

In the eligible RRMM cohort, the median duration of follow-up for all treated patients (which includes duration up to death for patients who died and duration up to last date known alive for surviving patients), was 10.2 months (range = 0.2 to 24.0). In the idecabtagene vicleucel cohort, the median duration of follow-up after idecabtagene vicleucel infusion was 11.3 months (range = 0.2 to 18.6). These numbers were current as of the October 30, 2019, data cut-off date of the KarMMa study. After a median follow-up time of 15 months for all surviving patients in the eligible RRMM cohort, 55.8% of patients had died, while after a median follow-up time of ||||| |||| for all surviving patients in the idecabtagene vicleucel–treated cohort, ||||||% of patients had died. The median OS for RWE patients in the eligible RRMM cohort was 14.7 months (|||||||||) and ||||||||||||||||||||||||||||||| in the idecabtagene vicleucel cohort. A comparison of the 2 groups yielded an OS hazard ratio (HR) of (||||||||||||||||||||), favouring the idecabtagene vicleucel cohort compared with the eligible RRMM cohort treated with available therapy |||||||||||.

The median TTR for responders was 1.1 months (range = 0.2 to 8.6) in the eligible RRMM cohort versus 1.0 month (range = 0.5 to 8.8) in the idecabtagene vicleucel cohort. As of the data cut-off date, the median DoR was 9.0 months (95% CI, 7.5 to ||||||) for 30.5% of responders in the eligible RRMM cohort versus 11.0 months (95% CI, ||||||||||||) for 73.4% of responders in the idecabtagene vicleucel cohort. The HR for DoR was |||||||||||||||||. After a median follow-up time of |||||||||||||||| for all RW patients in the eligible RRMM cohort, ||||| of patients had a PFS event (progressive disease [PD] or death), while after a median follow-up time of 11.3 months for all patients in the idecabtagene vicleucel cohort, ||||| of patients had a PFS event. A comparison between the 2 groups yielded a PFS |||||||||||||||||||||||||||||||||||||||||||, favouring the idecabtagene vicleucel cohort compared with the eligible RRMM cohort treated with available therapy (||||||||||).

The analysis using MAIC yielded similar results. In the treated population, idecabtagene vicleucel was more efficacious than conventional care in terms of the ORR (odds ratio = |||||||||||||||||||||||||) and also in the population using the target dose of 450 × 10⁶ CAR T cells. There was also a better PFS in the idecabtagene vicleucel–treated population compared with conventional care (HR = ||||||||||||||||||||||||||||||) and in patients receiving a target dose of 450 × 10⁶ CAR T cells (|||||||||||||||||||||||||||||||), and longer OS, with the population treated with idecabtagene vicleucel performing better than conventional care in the MAMMOTH treated population in the adjusted MAIC (HR = ||||||||||||||||) and target dose (||||||||||||||||||||||||||||||||).

Harms Results

No harms data were provided in any of the analyses submitted.

Critical Appraisal

The NDS-MM-003 study has several limitations. The missing data from the databases throughout the process used to create the eligible cohort, which implies potential bias due to the need to rely on multiple imputation methods, increasing the uncertainty in effect estimates. Data were missing for a number of variables which would have been useful to include in the propensity scores. Although later addressed in the propensity score balancing, baseline differences in multiple variables observed between the RRMM cohort, the eligible RRMM cohort, and the idecabtagene vicleucel cohort (e.g., the presence of plasmacytoma or refractory status) suggest the possibility of other potentially unmeasured confounders, and the prospect that these can be unevenly distributed between groups without proper randomization. Prior to the IPTW adjustment, with the exception of age, the imbalance of the prognostic factors and prior treatments suggests the eligible RRMM cohort was a less heavily pretreated and a less refractory population with a more favourable prognosis overall compared with the idecabtagene vicleucel cohort. Differences that arose during study design and when combining patients across databases (i.e., selecting patients from different databases) may result in the introduction of clinical heterogeneity that cannot be accounted for in propensity score modelling. Furthermore, investigators used, in part, a data-driven approach; clinical relevance should be considered to select the covariates used in propensity score models.

Although the population from the eligible RRMM cohort is similar to the idecabtagene vicleucel cohort in several measures, generalizability of this and the matched final cohort, as well as the comparators, can be a concern in Canadian clinical practice due to the different and varied therapies included in the regimens provided. The availability of such regimens and drugs will vary depending on funding and feasibility for each province and territory. Only a fraction of the regimens evaluated as comparators in this indirect comparison were mentioned by clinical experts consulted by CADTH as being commonly used in clinical practice in Canada.

The second comparative analysis (a MAIC) was limited to study-level aggregated data from the full-text publication for MAMMOTH and compared to the KarMMa study. Although the unanchored MAIC had an appropriate rationale for comparing individual patient data from a single-arm study with aggregate published data, it is not clear if the underlying assumption of the unanchored MAIC that all effect modifiers and prognostic factors have been accounted for or if all of the relevant differences in patient characteristics were captured, as many identified prognostic elements were not reported on in MAMMOTH. Furthermore, the MAIC cannot address heterogeneity in study designs, with possible differences in outcome assessment and definitions (e.g., differences in setting and follow-up) that generate uncertainty. Potential violations of the proportionality assumption were observed in the time-to-event outcomes. The different treatments obtained in the MAMMOTH study also generate uncertainty about the generalizability of the results, particularly from the Canadian perspective, due to the fact that approximately a third of the comparators are not widely available in the different provinces and territories.

Conclusions

Evidence from the single-arm, open-label KarMMa-001 study suggests that the CAR T-cell therapy idecabtagene vicleucel at a target dose of 450 × 10⁶ CAR T cells is associated with an induction of response, based on an ORR of 81.5% and a CRR of 35.2%. Results suggest that treatment with idecabtagene vicleucel may offer a benefit in terms of DoR, MRD, TTR, TTP, OS, and PFS, but none of these outcomes were controlled for multiplicity and all should be interpreted with consideration of type I error. The data for HRQoL outcomes were descriptive and could not be interpreted due to missing information. These effect estimates come from evidence with a risk of bias due to the lack of estimations from a comparison arm in a randomized design, and must be weighed against the observed undesirable effects, including AEs (such as CRS), the need for LDC, and bridging therapy. All patients treated with idecabtagene vicleucel reported at least 1 AE. The most commonly reported AEs were neutropenia, CRS, anemia, and thrombocytopenia.

Results of 2 indirect treatment comparisons (ITCs) of the population from the KarMMa study of individual patient observational (RWE) data and aggregated data from published studies add support for the beneficial effects observed in KarMMa, but also include a risk of bias due to inherent design differences in the bodies of evidence that cannot be adjusted for statistically, the potential impact of unmeasured and unaccounted-for prognostic factors and effect modifiers in the models, and undermined generalizability due to the inclusion of irrelevant comparators. Given the limitations of both ITCs and the absence of direct comparative evidence, any potential benefit of idecabtagene vicleucel over other treatment regimens used in this patient population remains unknown.

Introduction

Disease Background

Multiple myeloma is a malignant, clonal bone marrow plasma cell disorder that causes clinical features of renal failure, bone lesions, hypercalcemia, and bone marrow suppression resulting from excessive production of monoclonal proteins as well as direct tumour cell effects.^1,2

The second most common of hematologic malignancies, MM represents approximately 1% of all cancers and roughly 10% to 15% of all hematologic malignancies. It primarily affects older individuals and is extremely rare in patients younger than 30 years old.^3,4 Data from GLOBOCAN in 2012 estimated a global incidence of 114,000 patients and 80,000 deaths.¹⁴ In Canada, a recent epidemiological study¹ found the median age at diagnosis was 70.07 years (standard deviation = 12.08), with an average national incidence rate of 5.42 cases per 100,000 individuals per year, a steady rise in the annual rate of 0.96 cases per million individuals per year, and an estimated 3,400 new cases annually.^1,5 At the provincial level, Quebec and Ontario have the highest incidence rates of MM in Canada (5.82 and 5.66 cases per 100,000 individuals per year, respectively). The mean national mortality rate in Canada stands at 3.57 deaths per 100,000 individuals per year.¹

An initial diagnosis of MM usually occurs when patients are referred to specialists by primary care physicians or by encountering nonspecific symptoms (fatigue, bone pain) and other incidental findings. The diagnosis of symptomatic MM (myeloma that necessitates treatment) is made based on IMWG recommendations.¹⁵ Specifically, the patient must have a clonal bone marrow plasma cell level of greater than 10% and exhibit any 1 of the following: hypercalcemia, renal insufficiency, anemia, bone lesions, or a clonal bone marrow plasma cell level of at least 60%. Additional indicators of MM include a serum free light chain (FLC) ratio of at least 100 or multiple focal lesions on MRI studies. In recent years, a global trend of marked improvement in 5-year survival rates has been reported in patients with MM. However, despite these advances in treatment and prognosis, MM remains highly fatal and relapses are common.¹⁶

Patients are considered to have RRMM when disease recurs after prior response as determined by objective laboratory and radiological criteria.⁶ From there, patients begin moving to successive therapies as they advance in their course of the disease with further relapses and more lines of therapies. In patients who experience relapse and progression despite receiving several therapies, outcomes are usually poor, with infrequent CRs, a median PFS of 3 to 4 months, and a median OS of 8 to 9 months.⁷

Standards of Therapy

The goal of treatment of RRMM is to achieve disease control with acceptable toxicity and sufficient patient-defined HRQoL. This goal is reliant on effective systemic chemotherapy and supportive measures (pain control, antibiotics, kyphoplasty, radiation therapy, dialysis, and psychosocial supports).

Initial therapy for patients with MM depends on their eligibility for a bone marrow transplant. For patients who are eligible for a transplant the current treatment includes cycles of induction with conventional chemotherapy, proteasome inhibitors, and a corticosteroid; for example, this phase includes cycles of a combination of cyclophosphamide, bortezomib, and dexamethasone (CyBorD), or IMiDs such as a combination of lenalidomide, bortezomib, and dexamethasone. The induction phase is followed by a harvest of stem cells in preparation for an autologous stem cell transplant. After the transplant, consolidation therapy may be warranted, followed by maintenance therapy, typically lenalidomide or bortezomib. In patients who are not eligible for transplant the options include a novel IMiD, or proteasome inhibitors and a steroid. These can include lenalidomide, CyBorD, or a combination of lenalidomide, bortezomib, and dexamethasone.

For patients with RRMM, although new options, including recently approved drugs such as daratumumab and isatuximab,⁷ are emerging and increasingly available, no standard of care has been fully established, and the choice of the optimal strategy at relapse remains a challenge for clinicians. Currently, the options for patients with RRMM are limited to second to fifth lines of therapies that include several drugs, and the sequencing of treatments depends largely on the type of response to previous therapies (i.e., previous response to lenalidomide- or bortezomib-based therapies such as CyBorD).

Regimens commonly used in the Canadian landscape include combinations of daratumumab, bortezomib and dexamethasone; Vd; pomalidomide and dexamethasone; and carfilzomib and dexamethasone, all of which are regimens used when there is resistance to lenalidomide. Also, regimens used when there is resistance to bortezomib-based therapies include combinations of daratumumab, lenalidomide and dexamethasone; carfilzomib, lenalidomide and dexamethasone; lenalidomide and dexamethasone.^7-9

Drug

Chimeric antigen receptor T cells are modified T cells that express a CAR directed against specific antigens expressed only in tumours and not healthy tissues, potentially limiting toxicity. As these T cells are generated from autologous T cells collected from patients by leukapheresis, they represent an individualized therapeutic approach.^17,18

Idecabtagene vicleucel is a cell therapy formed by CAR T cells directed against the BCMA. It is indicated for the treatment of adult patients with MM who have received at least 3 prior therapies, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody, and who are refractory to their last treatment. The reimbursement request from the sponsor is according to the indication. Idecabtagene vicleucel is provided as a single-dose, 1-time treatment in 1 or more patient-specific infusion bags. Each single infusion bag of idecabtagene vicleucel contains a suspension of 275 × 10⁶ to 520 × 10⁶ CAR T cells. The target dose is 450 × 10⁶ CAR T cells within a range of 275 × 10⁶ to 520 × 10⁶ CAR T cells.¹⁹ Pretreatment includes the use of LDC with cyclophosphamide 300 mg/m² IV and fludarabine 30 mg/m² IV for 3 days. Idecabtagene vicleucel infusion is to be administered 2 days after completion of LDC. To minimize the risk of infusion reactions, acetaminophen (500 mg to 1,000 mg orally) and diphenhydramine 12.5 mg IV (or 25 mg to 50 mg orally), or another H₁-antihistamine approximately 30 minutes to 60 minutes are administered before the idecabtagene vicleucel infusion. Patients should be monitored in a specialized treatment centre at least daily for 7 days following IV infusions for signs of CRS, which may be fatal or life-threatening, and neurologic toxicities, which may be severe or life-threatening and may occur concurrently with CRS, after CRS resolution, or in the absence of CRS.

Table 3 lists the main characteristics of idecabtagene vicleucel.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on input provided by patient groups.

About the Patient Groups and Information Gathered

Myeloma Canada submitted the patient input for this review. Founded in 2005, Myeloma Canada is the only national charitable organization created by and for Canadians affected by MM. The organization works to improve the lives of those affected by myeloma by empowering the community through awareness, education, and advocacy programs, and supporting clinical research to find the cure.

Information from this input was gathered through a patient survey. The survey was accessed through email and social media from December 17, 2020, to January 4, 2021. The 388 individuals with myeloma who responded to the survey were divided into 2 sets of patients. Subset 1, with 161 respondents from across Canada, did not have experience with the treatment under review. However, they did qualify to receive the treatment based on previous therapy regimens. Subset 2 consisted of 14 respondents who had received a CAR T-cell therapy. These 14 respondents were from Ontario (36%), Alberta (21%), British Columbia (21%), Quebec (14%) and Saskatchewan (7%).

Myeloma Canada received external help from Colucci Health to summarize the data and help prepare this report.

Disease Experience

When respondents were asked to rate on a scale of 1 to 4 (1 = “not important”; 5 = “very important”) how important it is for them to control various aspects of their myeloma, 70% of the respondents indicated that their major concern was control of infections, followed by kidney problems (56%) and mobility (51%). Other reported outcomes were neuropathy (pain, numbness, tingling, swelling, or muscle weakness), fatigue, and pain.

When asked, “Please rate on a scale of one to five (one is ‘not at all’, and five is ‘significant impact’), how much symptoms associated with myeloma impact or limit your day-to-day activities and quality of life (QoL),” the respondents (n = 315) indicated that their symptoms significantly affected their ability to travel (35%), work (26%) and exercise (21%).

The patient input indicated that myeloma is a relapsing-remitting cancer; it alternates between periods of symptoms and/or complications that need to be treated, and periods of remission that do not require treatment. Providing patients with an effective treatment and giving them a prolonged remission with limited side effects can play a major role in their well-being and allow patients to live a full and productive life.

Experiences With Currently Available Treatments

When asked, “If you are taking a treatment or were to consider taking treatment for your myeloma, how important is it to improve your overall QoL? Rate on a scale of one to five, one is ‘not important’ and five is ‘extremely important’,” 61% of the respondents (n = 288) expressed that it is extremely important, 30% felt it was very important, and 9% felt it was important. Additionally, 40% of the respondents (n = 315) visit a cancer centre once a month.

Respondents (n = 313) identified drug costs (35%), parking costs (33%), and lost income due to absence from work (29%) as their most significant financial implications. Other economic factors included travel costs, medical supply costs, accommodation costs, and drug administration fees.

Respondents were asked, “Do you need the support of a family member or caregiver to help you manage your myeloma or your treatment-related symptoms?” and 54% (n = 314) answered “yes.”

The respondents (n = 266) identified effective treatments, fewer side effects, and better QoL as the most important aspects when it came to treating their myeloma. The following are examples of respondents’ explanations of what they considered to be the most important aspects of treating their disease:

• “Efficacy of the treatment for improved prognosis and ability to maintain similar quality and function of life activities.”

• “Extending my life and having hope for treatments going forward and being pain free.”

• “I would like to get off medications and as close to a long remission and cure as possible as I am a young myeloma patient.”

• “It’s important that I am able to receive the most up-to-date treatment and have access to the best minds in the field. Considerations for my mental health and how best to maintain a positive outlook is also very important.”

• “Being able to have a lifestyle rather than just a world revolving around treatment.”

• “Minimizing side effects and portability or reduced frequency of time in hospital.”

Improved Outcomes

Respondents From Subset 1

Of 297 patients, 161 had received at least 3 prior therapies, including an IMiD such as thalidomide, lenalidomide (Revlimid), or pomalidomide (Pomalyst), a proteasome inhibitor such as bortezomib (Velcade), carfilzomib (Kyprolis), or ixazomib (Ninlaro), and an anti-CD38 antibody, either daratumumab (Darzalex) or isatuximab (Sarclisa).

Seventy-three percent (n = 117) of respondents answered “yes” to having previously heard of CAR T-cell immunotherapy, 23% answered “no,” and 4% did not know what CAR T-cell therapy was. Respondents from subset 1 were asked to select the statement that best fit their understanding of CAR T-cell immunotherapy from the following options: “I don’t know,” “T cells (T lymphocytes, which are special types of white blood cells) are first harvested from a cancer patient and then modified in a laboratory before they are reintroduced into the patient’s body intravenously,” “CAR T-cell immunotherapy involves genetically modified cells responsible for killing cancer cells,” “CAR T-cell immunotherapy is a new type of blood cancer treatment,” and “All of the above.”

Fifty-seven percent of the respondents (n = 161) answered correctly by selecting “all of the above.” Myeloma Canada interpreted these results to mean more than half of respondents had a concrete understanding of the CAR T-cell therapy process and how it differs from their current treatment regimens.

When respondents were asked what they believe to be the benefits of CAR T-cell immunotherapy compared to the type of treatment they are currently receiving or have already received, the majority (75%) of respondents (n = 118) replied that they believed they would receive all of the benefits listed in the survey. Namely, prolonged remission, not having to be on treatment because they are in remission, better QoL, better overall health, better control of myeloma symptoms, and fewer side effects.

Respondents (n = 155) were asked, “Considering that treatment options are becoming limited at this point in your myeloma journey, other than participating in clinical trials or another treatment, indicate what an estimated prolonged remission of 34.2 months means to you at this stage in your life.” A prolonged remission of 34.2 months was “extremely desirable” to 89 respondents, “very desirable” to 19 respondents, “desirable” to 16 respondents, “somewhat desirable” to 4 respondents, and 25 respondents did not know what it meant to them.

Myeloma Canada explained the treatment procedures and risks to the respondents as follows: “In some cases, the patient’s myeloma can progress during this time (i.e., the myeloma comes back) and the patient may require an additional (bridging) therapy. If this happens, the patient may no longer be eligible to proceed to the next step which is to receive their modified T cells. In this case, other treatment options are available and will be presented to the patient.” Respondents were asked to indicate, from a list, which sentence best reflects their sentiments about this kind of situation. Forty-six percent (n = 72) of the 156 respondents said they were not sure how it makes them feel and would like to discuss this situation further with their hematologist or oncologist. Twenty-three percent of respondents (n = 36) were willing to take the risk that their myeloma would return. Survey respondents reported that a loss of or decrease in cognitive abilities (66%) and loss of physical abilities and mobility (54%) were the most concerning CAR T-cell treatment side effects. Respondents were asked to rate side effects on a scale from 1 (most troublesome) to 5 (least troublesome). Among other concerning side effects reported by respondents were anemia, neutropenia, and thrombocytopenia.

When respondents (n = 151) were asked if they felt CAR T-cell immunotherapy could improve their health and well-being, 48% answered “yes,” 4% answered “no,” and 48% did not know. Additionally, 58% of respondents (n = 150) replied that they expected CAR T-cell immunotherapy could improve their long-term health outlook. When asked, “With what you know today, what treatment option would you consider first as your next treatment?,” 35% of the 151 respondents would choose CAR T-cell immunotherapy, 15% would choose another treatment suggested to them, 9% would choose a clinical trial, and 42% did not know what they would choose.

The following quotes express respondents’ feelings toward CAR T-cell immunotherapy:

• “The potential of extending my life for shy of three years is much better than going through clinical trials that are only effective on average for 9 months.”

• “The patient and caregiver are under a lot of stress try[ing] to adjust. Adjusting to travel, find[ing] accommodation, finding your way to the hospital, new doctors, having to go to more than one hospital. Unknown side effects. More medical expenses.”

• “3 & a half years [of prolonged remission] is not that long.”

• “I have been in four clinical trials. I would try another before I would consider CAR T-cell immunotherapy if I were eligible. I am 80 years old.”

Experience With Drug Under Review

Respondents From Subset 2

Fourteen respondents have been treated or are in the process of receiving a CAR T-cell immunotherapy to treat their myeloma. These respondents were asked which CAR T-cell therapy they received. Only 1 respondent received the treatment under review, 5answered they had received another CAR T-cell therapy, and 8 did not know which CAR T-cell therapy they had received. Myeloma Canada presented the data on all the patients who had received a CAR T-cell therapy, providing a directional perspective of this new therapy in an attempt to give CADTH a better understanding of the overall patient experience.

When respondents were asked, “How long ago did you receive your idecabtagene vicleucel CAR T-cell therapy?” 1 respondent answered, “less than three months ago,” 2 answered “between three and six months ago,” 1 answered “between six and 12 months ago,” 1 answered “over a year ago,” and 1 answered “over two or more years ago.” In addition, respondents (n = 6) were asked, based on their experience of the idecabtagene vicleucel CAR T-cell immunotherapy, how they would rate the effectiveness of this treatment in helping to control their myeloma. Among the 6 respondents, 1 answered it was “extremely effective,” 3 answered “very effective,” 1 answered “effective,” and 1 answered “not effective.”

Three of the 6 respondents answered that the overall side effects from idecabtagene vicleucel CAR T-cell immunotherapy were “very tolerable,” 2 answered they were “tolerable,” and 1 answered they were “somewhat tolerable.” Among the most frequent side effects seen in patients who participated in the idecabtagene vicleucel CAR T-cell immunotherapy trial, respondents (n = 6) reported neutropenia (n = 4) and decrease of cognitive abilities (n = 4) were the most bearable side effects. Other side effects considered most bearable by respondents were loss of physical abilities (n = 3), neuropathy (n = 3), pain (n = 3), muscle weakness (n = 3) and typical cold-like symptoms (n = 3).

Respondents were asked about the importance of not being on a treatment for their myeloma after idecabtagene vicleucel CAR T-cell immunotherapy, either for a prolonged or indeterminate period. Four of the 6 respondents answered it was “extremely important,” 1 answered “somewhat important,” and 1 answered, “not important.”

Five of the 6 respondents agreed that CAR T-cell therapy (idecabtagene vicleucel) met their expectations in treating myeloma. Quotes indicating respondents’ expectations included “I am so far in very, very good remission,” “I was in remission for about a year,” “Didn’t get depth of response as we were hoping,” “No more chemo!,” and “The required 2 weeks in hospital seemed very, very long.”

All 6 respondents agreed that CAR T-cell therapy (idecabtagene vicleucel) improved their long-term health outcomes. Three responded with “Yes, I’m still alive!,” “I would otherwise have run out of options, but more are now available,” and “I had it done 2 times and it gave me drug free time. I would do it again if they would allow me.”

Of the 6 respondents, 3 are still in remission, 1 indicated that their myeloma had just come back and they are about to start a new treatment soon, and 2 stated that their myeloma did come back and they are currently under a new treatment regimen.

Additional Information

Myeloma Canada highlighted that patients are looking for effective treatments. Given that no single treatment is effective for all myeloma patients, various treatment options must be available to improve patients’ prognoses and QoL. As represented in the patients’ responses, a CAR T-cell therapy can provide an important benefit for patients despite the number and severity of the side effects. Even though only a small number of respondents had experience with the treatment under review, about half of the respondents fit the criteria. For these patients, treatment options are becoming limited. Unfortunately, their lives depend on these treatment options because there is still no cure for myeloma.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in 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, providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy. In addition, as part of the idecabtagene vicleucel review, a panel of 4 clinical experts from across Canada was convened to characterize unmet therapeutic needs, help identify and communicate situations in which there are gaps in the evidence that could be addressed through the collection of additional data, promote the early identification of potential implementation challenges, acquire further insight into the clinical management of patients living with a condition, and explore the potential place in therapy of the drug (e.g., potential reimbursement conditions). A summary of this panel discussion follows.

Unmet Needs

The clinical experts agreed that not all patients with MM respond to available treatments, and they eventually become refractory to all treatment options. These patients have a short life expectancy, and their remaining life is often highly medicalized. Treatments are needed to control the disease and provide patients with a break from therapy. The experts stated that there are few options after third- and fourth-line therapies, and the only standard options available after 3 therapies are palliation and supportive care. The majority of CAR T-cell treatments are financially burdensome either at the individual patient level or for the public health care system. Ideally, treatments should be affordable on a large scale.

Place in Therapy

The clinical experts consulted by CADTH acknowledged that idecabtagene vicleucel is a therapy with a different mechanism of action and therefore is appropriate for patients who have already been exposed to standard therapies and for whom continuing with standard therapies is not expected to have any further benefit. This treatment should only be given as monotherapy and not combined with other active agents. Formal clinical trials would be required to establish if this treatment can be complementary to other treatments. Based on current evidence, the clinical experts indicated that treatment should be used after patients become triple-exposed or double-refractory (i.e., following failure with or intolerance to lenalidomide and/or daratumumab and bortezomib-exposed). Treatment after patients become triple-exposed or double-refractory is inevitable as most other treatment options continue to be moved up into the front line. In the current treatment paradigm, CAR T-cell therapy would be well positioned in the third line (current Canadian approval) and fourth line. The clinical experts noted that, should daratumumab, bortezomib, lenalidomide, and dexamethasone be approved as a front-line therapy, CAR T cells could become a second-line option.

However, 1 of the clinical experts argued that it is unlikely CAR T cells will cause a shift in the current treatment paradigm as it is being considered as a last-line therapy, and it falls in line after current therapies. Given the toxicity of idecabtagene vicleucel, the expert stated that it is unlikely to move up in the treatment paradigm until more clinical trials are completed, and even then, only if superior efficacy compared with current standard therapies is demonstrated. The expert added that, while idecabtagene vicleucel may be tested in combination with other agents, those trials are far from complete.

Patient Population

The eligible population may be small because many patients already reach a palliative stage by fourth-line treatment due to rapid disease progression, age, or frailty and toxicities from prior treatments. Those patients who meet the inclusion criteria for the phase II trial would be the most appropriate and have the greatest unmet need. Patients with a good performance status and those who have been previously treated with or are refractory to proteosome inhibitors, IMiDs, and anti-CD38 monoclonal antibodies have an unmet need. Worse outcomes tend to occur if patients in rural communities forego important treatments just to stay closer to home and if patients with limited financial means are unable to pay for appropriate supportive care.

Clinical experts stated that the patients best suited for this treatment would be those with an ECOG status of 0 or 1, good organ status, and a willingness to travel (as this therapy will not be offered in smaller centres). Those with relapsed and refractory disease should be prioritized first. Younger, fit patients with the least number of prior therapies (3 prior lines) are most likely to benefit. Because idecabtagene vicleucel requires a certain level of fitness, the presence of comorbidities and fitness are important aspects to consider when determining whom to prioritize. Patients eligible for this novel therapy would need to fit as closely as possible the eligibility criteria of the phase II trial that is under review. If capacity limitations exist, then the relevant subgroups with the best response to this new agent should be assessed.

The clinical experts agreed that patients best suited for idecabtagene vicleucel would be identified using the inclusion and exclusion criteria from the study, and that clinical judgment should be based on prior therapies, patient comorbidities, and fitness. Clear, transparent, and evidence-based guidelines are important because capacity for treatment will likely be greater than the need. All standard diagnostic and baseline tests are available broadly in Canada. Once patients with relapsed or refractory disease are recognized, it is recommended that health regions and/or provinces form adjudication committees that can provide a recommendation about patient eligibility for this treatment to remove bias as to which patients may be eligible for this treatment.

Patients the least suitable for treatment with idecabtagene vicleucel would be those with a poor performance status and those with other options available with a reasonable duration of disease control. Additionally, patients who are frail, elderly, or have significant renal failure, neurologic disease, and an ECOG status of at least 2 related to myeloma would be the least suitable candidates for idecabtagene vicleucel. In general, the patients least suitable for treatment would be those who do not resemble the eligibility criteria of the pivotal trial under review.

Assessing Response to Treatment

The clinical experts agreed that a PFS of longer than 6 months and an OS of greater than 1 year without life-threatening toxicities would be clinically meaningful responses to treatment. They stated that the standard assessment of treatment response would be appropriate. Standard assessment would be conducted every month using serum protein electrophoresis or FLC testing for at least 1 year, after which assessments would be less frequent. However, there currently are no published data on re-treatment with this drug. Should published data indicate that re-treatment with this drug is efficacious, then it should be considered on a case-by-case basis when sufficient supplies of cryopreserved drugs are available.

Discontinuing Treatment

The clinical experts agreed that discontinuing CAR T-cell treatment is not applicable because once therapy is initiated, it is not able to be modified, stopped, or reversed. It is a 1-time infusion of cells; hence, discontinuation of treatment is difficult to perform as the effect is largely irreversible once administered. However, patients who become unwell before infusion may be deemed too sick to proceed with infusion.

Prescribing Conditions

Idecabtagene vicleucel must be administered in a setting supervised by specialists trained in bone marrow transplants, or a tertiary centre that is familiar with complications related to stem cell transplants or is able to accommodate phase I trials, with adequate ICU, emergency, neurology, and infectious disease departments. This will limit availability for some patients. In Canada these centres are those that are administering blood marrow transplants at institutions across the country accredited by Health Canada and the Foundation for the Accreditation of Cellular Therapy.

Clinical experts agreed that a specialist is required to diagnose, treat, and monitor patients who receive idecabtagene vicleucel. It would require a hematologist or oncologist with specific experience in cellular therapy; the majority of these experts would also have stem cell transplant experience.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

Two clinician groups provided input on the reimbursement review of idecabtagene vicleucel for the treatment of adult patients with MM who have received at least 3 prior therapies, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody.

The CMRG is a charitable organization of myeloma physicians from 22 academic centres in Canada. The CMRG’s main purposes include conducting investigator-initiated academic clinical trials for myeloma patients, maintaining a national myeloma database, and generating consensus statements for myeloma management. The input provided to CADTH for the submission of idecabtagene vicleucel was drafted by the CMRG’s chief medical officer and sent to all members for input. The clinicians who agreed with and commented on the input were included as part of the input.

The Ontario Health Hematology Cancer DAC provides evidence-based clinical and health system guidance on drug-related issues in support of CCO’s mandate, including provincial drug reimbursement programs and the systemic treatment program. For the review of idecabtagene vicleucel, the Ontario Health DAC collected information via email.

Unmet Needs

The Ontario Health Hematology Cancer DAC noted that the current treatment goal for patients is to prolong life, delay disease progression, and control disease. The CMRG added that controlling disease manifestations (bone destruction and pain, renal failure, hypercalcemia, and low blood counts), maintaining the control of the myeloma and its manifestations (PFS), minimizing adverse effects and caregiver burden, and optimizing QoL are current treatment goals.

The DAC noted that currently available treatments have low response rates and short durations of response. The CMRG added that myeloma remains incurable, and patients will eventually become refractory to all available agents. The CMRG also added that about 15% of patients have extremely aggressive disease at diagnosis and die within 1 to 2 years despite all therapies. In addition, the group noted that many patients will experience increasingly aggressive disease with time; at some point, patients succumb to their disease and often have a miserable QoL in the end stages. The CMRG noted that some treatments are not publicly funded despite Health Canada approval, and that in some instances only privately insured patients can access these potentially life-extending therapies, which poses an ethical challenge for patients, caregivers, and health care providers.

Patients with the greatest unmet need according to the DAC are those whose life expectancy is limited, including triple-exposed patients. The group noted that the drug under review would address this unmet need. However, according to the clinician groups, there may be challenges in delivering this drug in patients with comorbidities and low blood counts.

The CMRG added that the greatest unmet need is for patients who have failed therapy with proteasome inhibitors, IMiDs, and anti-CD38 monoclonal antibodies, as no effective therapy is available, and that the drug under review would address this unmet need.

Place in Therapy

Both the Ontario Health DAC and the CMRG noted that, as per the clinical trial, the drug under review would be most appropriate for patients who have been triple-exposed to MMRM therapies. The CMRG noted that the drug addresses the underlying malignancy and would be used in sequence after the other lines of therapy discussed earlier in this summary.

The DAC noted that there is insufficient information to determine whether it would be appropriate for patients to try other treatments before idecabtagene vicleucel. They cautioned that the risk of treating patients with other regimens such as Cd may result in organ dysfunction or treatment-related toxicities that could make the delivery of CAR T-cell therapy difficult. The CMRG added that no other effective drugs are available other than idecabtagene vicleucel in the fourth-line setting.

With respect to sequencing, the Ontario Health Hematology Cancer DAC noted that there is no sequencing information available other than that from the triple-exposed population. The CMRG added that, after failure of idecabtagene vicleucel, patients would most likely only be offered palliative radiotherapy and other supportive measures. The group noted that all effective and/or funded regimens have already been used. The CMRG added that many clinical trials of new immunotherapies exclude patients who have received anti-BCMA immunotherapy (such as idecabtagene vicleucel) from receiving another agent targeting BCMAs. Because BCMAs are the most common targets of new immunotherapeutic platforms, subsequent trial participation would be limited.

Patient Population

The Ontario Health DAC noted no information is available to identify the patient subgroups most likely to respond to treatment. The DAC added that the clinical trial recruited patients with a good performance status.

The CMRG noted that patients with other poor prognosis factors, such as extramedullary disease and experience with high-risk cytogenetics, do not fare significantly worse, and older age does not appear to be an exclusion factor. The group added that, from a practical point of view, patients whose disease is “progressing at a rate anticipated to allow them to remain stable and relatively well during the 4-5 week waiting time for CAR T-cell processing would be best suited for this treatment in order to avoid significant attrition.”

The DAC noted that patients best suited for the treatment should be identified through BCMA staining, which is not currently part of routine testing. They added that patients can be identified by applying routine criteria for myeloma relapse or progression. The CMRG added that identifying patients who should be treated with the drug under review will require careful matching of available resources with the potential numbers of patients with relapsed myeloma. Idecabtagene vicleucel patients would require an inpatient bed for approximately 2 weeks (or longer if complications occur) and may require readmission. The CMRG added that these patients require potentially expensive ancillary measures to treat CRS, neurotoxicity, and infections, and may need ICU support. Specialized training is required for staff, and each medical centre’s infrastructure and clinical pathways must be specific to meet the safety standards for treatment and follow-up. Appropriate guidelines need to be established.

The DAC noted that patient selection should align with the inclusion and exclusion criteria of the trial for treatment; those that do not align are least suited. The CMRG added that frail patients, those with rapidly proliferating disease, ongoing infection, and significant organ dysfunction, and patients with pre-existing pancytopenia would be least suitable for treatment with the drug under review.

The DAC added that no data are available to predict which patients are most likely to exhibit a response. However, the CMRG noted that patients with a good performance status and organ function, minimal or no comorbidities, robust blood counts, low tumour burden, and indolent disease are most likely to have the best outcomes. The CMRG added that, given the similar response and PFS rates in subset analysis of the clinical trials for the drug under review, it will likely be easier to develop exclusion criteria than inclusion criteria.

Assessing Response to Treatment

According to the Ontario Health DAC, outcomes that are used to assess whether a patient responding to treatment in clinical trials should be the standard myeloma response criteria. The CMRG added that responses are based on the monoclonal protein marker in the serum and/or urine, bone marrow biopsy, and in some instances imaging studies. The CMRG also added that these align with the criteria used in clinical trials and include the emerging parameter of marrow MRD.

To assess clinically meaningful responses, the DAC noted that at least a partial response (PR) should be present. As per myeloma response criteria, such responses include improvement in blood counts, improvement in end organ function (i.e., renal function), reduction in skeletal events (i.e., bone pain or fractures), and overall improvement in QoL. The CMRG noted that clinically meaningful responses usually correlate with at least a partial remission as defined by IMWG consensus criteria. The CMRG added that these include improvement in symptoms (cessation of bone destruction with less pain, fractures, and need for radiotherapy), energy levels, and ability to perform daily activities.

According to the DAC, assessment for responses should be conducted every 1 to 2 months. However, the CMRG noted that, in myeloma, responses are generally assessed every 2 to 3 months, depending on clinical stability.

Discontinuing Treatment

Both the Ontario Health DAC and the CMRG noted that discontinuation criteria for the treatment is not applicable as this is usually a single treatment.

Prescribing Conditions

The DAC noted that delivery of the drug under review will be at tertiary hospitals and transplant centres with expertise in cellular therapy. The CMRG added that this therapy is appropriate for a major medical facility with expertise in other cellular therapies for hematologic malignancies. The group noted that close interactions between a specialized inpatient service, an ICU familiar with immunosuppressed cancer patients, and an outpatient facility experienced in handling urgent hematologic problems are necessary. The CMRG noted that appropriate coordination with the emergency department to expedite care of patients recently discharged following treatment with the drug under review is also required.

Drug Program Input

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

Jurisdictional Context

Relevant Comparators

The PAG noted that the choice of comparator in the submitted trial would need special consideration as there is no clear standard of care (i.e., when reaching the fourth line of therapy). The sponsor used several comparators based on the real-world NDS-MM-03 trial (i.e., pomalidomide plus dexamethasone, pomalidomide plus bortezomib plus dexamethasone, Cd, carfilzomib plus cyclophosphamide plus dexamethasone, Vd, and CyBorD). At this point, fourth-line treatment options would be cyclophosphamide plus dexamethasone, Vd, or CyBorD (which are funded by most jurisdictions). Although pomalidomide plus bortezomib plus dexamethasone was recently reviewed by CADTH with a reimbursement with conditions recommendation, it is not yet funded. Pomalidomide plus dexamethasone, Cd, and cyclophosphamide plus dexamethasone are typically not funded, although a few jurisdictions may consider them on a case-by-case basis. The clinical experts consulted by CADTH agreed with these assertions.

Policy Considerations for Reimbursing the Drug

Considerations for Initiation of Therapy

The PAG posed some questions related to decision-making, such as:

• What patient (or disease) characteristics would be important in determining who should be treated with this therapy?

• If capacity limitations exist, how should medical staff prioritize which patients are offered idecabtagene vicleucel?

• Do patients need to have disease considered refractory to an IMiD, a proteasome inhibitor, and an anti-CD38 antibody to receive idecabtagene vicleucel?

The characteristics of patients who should receive idecabtagene vicleucel are described in the clinician input section of this report. The clinician groups also stated that any patient with prior treatments with IMiD, proteasome inhibitor, or anti-CD38 antibodies should be allowed to receive treatment with idecabtagene vicleucel. The only unresolved concern was the exact line of therapy to recommend when initiating CAR T-cell treatment, as many patients could be considered refractory at different lines of therapy (i.e., second to fifth lines). They also noted that prioritization of the therapy would be a rare concern and should be made on a case-by-case basis; the same conclusion was drawn for the re-treatment considerations.

Considerations for Prescribing

The PAG identified areas associated with prescribing idecabtagene vicleucel, most notably that access would be limited to jurisdictional capacity. Given that the sponsor plans to establish 12 sites across Canada, out-of-province care may be needed for proper administration of the drug under review. Also, specialized centres need to be trained and accredited by the sponsor, limiting the availability of clinics where the therapy can be provided. There is a high human resource burden to obtain and maintain certification (including developing various protocols and supporting yearly audits). Similarly, access and prolonged stay in (or near) specialized centres, particularly for patients from remote areas, are needed, as is the ability to coordinate patient care and product preparation with an external organization or manufacturer. Several manufacturers may provide CAR T-cell therapies and there is an administrative burden on specialized sites to manage various protocols and systems for preparation and delivery of each product type.

Special Implementation Issues

Generalizability

The PAG noted that there may be potential for “indication creep” to earlier lines of therapy if, for example, triple-class exposure occurs in the first 2 lines of therapy. The PAG noted that the current sponsor request is for the use of idecabtagene vicleucel after 3 prior therapies, and the provisional algorithm and the budget impact analysis includes a scenario to use idecabtagene vicleucel in a third-line setting. Although the economic submission is based primarily on the KarMMa trial (≥ 3 prior regimens), the sponsor included KarMMa-3 (inclusion criteria of at least 2 but no greater than 4 prior MM regimens). It was therefore considered important to establish what would be the line of therapy (e.g., third or fourth) for idecabtagene vicleucel and which patient populations would be outside the scope of the reviewed indication. Furthermore, idecabtagene vicleucel may change the place in therapy of other (comparator) drugs and in subsequent lines.

Care Provision Issues

The PAG highlighted the adjacent use of resources with idecabtagene vicleucel patients, including leukapheresis, cell processing, and use of bridging and LDC. The management of adverse effects will also require indirect costs for nursing, hospital beds, ICUs, and administration of tocilizumab to treat CRS.

The trial protocol required bridging therapy (e.g., daratumumab), which may not be publicly funded. Clinical experts consulted by CADTH agreed that it would be important to discuss the impact and feasibility of this strategy in real-world practice.

According to the experts, there will be no need for BCMA testing (which was an exploratory objective in the trial), although they agreed that there are few data on which to base decision-making with high certainty.

No items were identified by PAG in relation to considerations for renewal or discontinuation of idecabtagene vicleucel therapy.

Clinical Evidence

The clinical evidence included in the review of idecabtagene vicleucel is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the literature that met the selection criteria specified in the review. The third section is intended to include sponsor-submitted, long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review. No additional evidence was identified for this review.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of idecabtagene vicleucel for the treatment of adult patients with MM who have received at least 3 prior therapies, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 4. Outcomes included in the CADTH review protocol reflect those considered to be important to patients, clinicians, and drug plans.

The systematic review protocol was established before the granting of a Notice of Compliance from Health Canada.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the Peer Review of Electronic Search Strategies checklist (https://www.cadth.ca/resources/finding-evidence/press).²⁰

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

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

The initial search was completed on January 21, 2021. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee on May 13, 2021.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist (https://www.cadth.ca/grey-matters).²¹ Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. In addition, the sponsor of the drug was contacted for information regarding unpublished studies.

Findings From the Literature

Of the 70 citations identified in the literature search, none were relevant according to the inclusion criteria and all were excluded, while 10 potentially relevant full-text reports from other sources were retrieved for scrutiny. One report describing 1 study is included in the CADTH systematic review. (Figure 1).

The included studies are summarized in Table 5. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of Studies

The KarMMa study was a phase II, open-label, multi-centre, single-arm trial conducted in 20 sites in the US, Spain, Italy, Germany, France, and Belgium, and 1 site in Canada. The study was run from January 3, 2018, when the first patient was enrolled, to the last data cut-off date of October 16, 2019. The main objective of the study was to evaluate the efficacy, defined as the ORR of idecabtagene vicleucel in patients with RRMM. Secondary objectives included the assessment of safety of idecabtagene vicleucel and other measures of efficacy and HRQoL, as described in Table 5. The study consisted of 3 periods:

• pretreatment (screening, leukapheresis, and bridging therapy, if needed)

• treatment (LDC and idecabtagene vicleucel infusion)

• posttreatment (post–idecabtagene vicleucel infusion).

The overall design of the study in the pretreatment and treatment periods is depicted in Figure 2.

Figure 2: Pretreatment and Treatment of Patients During the Study

ICF = informed consent form; Ide-cel = idecabtagene vicleucel; LD = lymphodepletion.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

The pretreatment period consisted of the screening process, during which patients were evaluated for their suitability to enter the study. Patients who accepted and were enrolled underwent leukapheresis to enable idecabtagene vicleucel product generation, manufactured on a per-patient basis. At the clinical investigator’s discretion, patients could receive bridging therapy for myeloma control following leukapheresis while idecabtagene vicleucel was being manufactured, as long as the last dose of bridging therapy was administered at least 14 days before the initiation of LDC. Bridging therapies could include corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies as single agents or in combination.

The treatment period began with the LDC. Patients who were eligible received a 3-day cycle of LDC with fludarabine and cyclophosphamide beginning 5 days before the target idecabtagene vicleucel infusion date. When LDC was completed, the idecabtagene vicleucel infusion was administered within a dose range of 150 × 10⁶ to 540 × 10⁶ CAR T cells per infusion. Before proceeding with idecabtagene vicleucel infusion, the patient’s clinical status must not have significantly worsened. The maximum dose (540 × 10⁶ CAR T cells) was defined by the upper limit of 450 × 10⁶ CAR T cells plus 20%. Patients were required to be hospitalized from the day of idecabtagene vicleucel infusion through day 14 post-infusion to monitor the risk of developing CRS and neurotoxicity.

After infusion with idecabtagene vicleucel, patients began the posttreatment period. In this safety phase, idecabtagene vicleucel expansion and persistence, HRQoL, myeloma response, disease status, and other end points were assessed at prespecified time intervals for all patients and will continue to be assessed for a minimum of 24 months from idecabtagene vicleucel infusion or until documented disease progression, whichever is longer. Patients with PD within 24 months of idecabtagene vicleucel infusion remain on the study and have select assessments collected until month 24. Patients who remain progression-free will have assessments collected for up to 5 years.

Re-treatment with a second infusion of idecabtagene vicleucel, including a second course of LDC, with or without bridging therapy, was allowed after disease progression if protocol-specified eligibility (inclusion and exclusion) criteria were met.

Amendments and Changes in the Conduct of the Study

Four protocol amendments were implemented during the conduction of the KarMMa study. The original protocol was amended 4 times before the data cut-off date of October 16, 2019. The first amendment modified the overall safety management plans in response to a then-recent SAE (a grade 4 neurotoxicity event) that occurred in the idecabtagene vicleucel phase I study. Amendment 2.0 was written to increase the idecabtagene vicleucel dose range, increase the sample size, modify the idecabtagene vicleucel overdose definition, and to incorporate feedback from health authorities. Amendment 3.0 provided further guidance on intercurrent illness or toxicity that is considered to place the subject at undue risk of proceeding to idecabtagene vicleucel infusion and for which idecabtagene vicleucel infusion should be delayed. Amendment 4.0 changed the MRD assessment by EuroFlow from a secondary objective and end point to an exploratory objective and end point. Specific points for the amendments to the protocol are summarized in Table 6.

Populations

Inclusion and Exclusion Criteria

Key inclusion and exclusion criteria of the KarMMa study are listed in Table 5. Briefly, the KarMMa study included patients with a diagnosis of RRMM who had received at least 3 prior antimyeloma therapy (AMT) regimens, were refractory to the last regimen received (defined as documented progressive disease during or within 60 days of completing treatment with the last antimyeloma drug regimen), and had been previously treated with an IMiD, a proteasome inhibitor, and an anti-CD38 antibody. Induction with or without a hematopoietic stem cell transplant and with or without maintenance therapy was considered a single regimen. Patients with solitary plasmacytomas, previous allogeneic stem cell transplant, relevant central nervous system (CNS) pathology, inadequate bone marrow function, and liver, renal, and other hematologically relevant conditions were excluded.

In addition, given that enrolled patients had RRMM and were refractory to their last line of therapy, it was expected that hematologic and organ function could worsen during the window between leukapheresis and initiation of LDC. To be eligible for LDC, antimyeloma bridging therapy must have been stopped 14 days before start of LDC, and patients were required to have adequate bone marrow, hepatic, and renal function, no infections or intercurrent illnesses, and adequate international normalized ratio and partial thromboplastin time.

Baseline Characteristics

Overall, key baseline characteristics of the enrolled population (n = 140) were broadly similar to those of the rest of the populations treated with idecabtagene vicleucel. Median ages in treatment groups were in a range of 54 years to 62 years, with a median of 60.5 years in the enrolled population; only a minority of all patients (3.6%) were in the group older 75 years of age. Most patients were male (58.6%) and White (80.7%), with similar distributions in the treated populations. The ECOG performance status was also similar between treatment populations, with the majority of patients established at an ECOG status of 1 (53.1% in the total population) and only 3 patients with a status of 3. All demographic and baseline characteristics for the enrolled population and the population treated with idecabtagene vicleucel are provided in Table 7.

The median duration of the disease was about 6 years in the enrolled and treated population, which is expected in patients with RRMM who have undergone several prior treatment regimens, according to the clinical experts consulted by CADTH. More than half of patients had been treated with 7 or more antimyeloma regimens, and more than 90% had received a stem cell transplant and radiation therapy. Most patients enrolled in the KarMMa study were double-refractory (in this case, more than 87% to an IMiD and proteasome inhibitor), and 26% of the enrolled population were penta-refractory. Most patients (> 85%) received bridging therapy while idecabtagene vicleucel was manufactured.

Interventions

After informed consent was obtained and screening eligibility was met, patients were enrolled and underwent leukapheresis to enable production of idecabtagene vicleucel. Idecabtagene vicleucel was manufactured on a per-patient basis and there was expected variability in the number of CAR T cells that were manufactured for each patient. Patients who had a product manufacturing failure or a product manufactured with a dose of less than 150 × 10⁶ CAR T cells could proceed to a second leukapheresis.

When necessary, patients could have received bridging therapy for myeloma disease control following leukapheresis while idecabtagene vicleucel was being manufactured (Figure 2). Bridging therapies could have included corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies as single agents or in combination. The decision of whether or not to initiate bridging therapy, given patients’ expected rate of deterioration while idecabtagene vicleucel was being manufactured, was left to each respective investigator. Experimental agents and myeloma therapies to which the patient had not been previously exposed were not to be used as bridging therapy. Per-protocol, antimyeloma bridging therapies were stopped at least 14 days before the start of LDC.

The treatment period began with LDC. Eligible patients received a 3-day cycle of LDC with cyclophosphamide (300 mg/m² IV infusion over 30 minutes) and fludarabine (30 mg/m² IV infusion over 30 minutes administered immediately after cyclophosphamide) both at days −5, −4, and −3 of the study. After completing the LDC, patients were infused with idecabtagene vicleucel. Prior to proceeding with idecabtagene vicleucel infusion, the patient’s clinical status must not have significantly worsened (with worsening defined as having a suspected or active infection, fever unrelated to MM, requirement of supplemental oxygen, cardiac arrhythmia, hypotension, or other non-hematological disorder, or taking medications prohibited in the protocol).

The selection of the idecabtagene vicleucel dose range was informed by evolving data from the phase II (CRB-401, part A) study. Doses ranging from 50 × 10⁶ to 800 × 10⁶ CAR T cells were explored. Based on an overall benefit-risk assessment, a safety review committee recommended a dose range of 150 × 10⁶ to 450 × 10⁶ CAR T cells to evaluate dose expansion. Due to an updated manufacturing process for idecabtagene vicleucel, the selected dose range for initiation of the MM-001 study was modified to 150 × 10⁶ to 300 × 10⁶ CAR T cells, with the intent of treating 80 patients. After the initial 4 patients were treated at the target dose of 150 × 10⁶ CAR T cells in the MM-001 study, subsequent patients were treated at the target dose of 300 × 10⁶ CAR T cells. Based on an emerging dose-response relationship from the CRB-401 study and acceptable ongoing safety at both the 150 × 10⁶ and 300 × 10⁶ CAR T cells target dose levels in MM-001, the dose range in MM-001 was modified from 150 × 10⁶ to 450 × 10⁶ CAR T cells, along with expansion of the total sample size, with the intent of targeting the remaining patients at 450 × 10⁶ CAR T cells. The maximum dose of idecabtagene vicleucel (540 × 10⁶ CAR T cells) was defined by the upper limit of 450 × 10⁶ CAR T cells plus 20%. The target dose was modified to 450 × 10⁶ CAR T cells.

Patients were pre-medicated approximately 30 minutes before idecabtagene vicleucel infusion; these pre-medication included acetaminophen 650 mg orally and diphenhydramine 12.5 mg IV or 25 to 50 mg orally (or equivalent). Corticosteroids, as pre-medication, were not allowed.

Bridging therapy was considered for any patient, given their expected rate of deterioration, while idecabtagene vicleucel was being manufactured, based on the clinical judgment of the respective investigator. Bridging therapies could have included corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies as single agents or in combination.

After idecabtagene vicleucel infusion, all patients were closely monitored for at least 14 days as inpatients for signs and symptoms of CRS or neurotoxicity post–idecabtagene vicleucel infusion. Tocilizumab was available at each site before infusion of each patient. Study sites had ICUs with the ability to manage CRS, neurotoxicity, and other potential complications.

Prohibited medications included systemic corticosteroids (unless used for CRS or neurotoxicity), any systemic MM therapy within 14 days before leukapheresis, any experimental agents for MM, bridging therapies not included in the protocol, any systemic MM therapy, including experimental agents, within 14 days of LDC, any concurrent chemotherapy, immunotherapy, biologic, experimental, or hormonal therapy following idecabtagene vicleucel infusion (posttreatment period) for treatment of MM before documentation of PD.

Subsequent AMTs were summarized for the idecabtagene vicleucel population and were also considered in the outcome time to subsequent AMT in the KarMMa study. In this instance, the idecabtagene vicleucel re-treatment (including LDC and bridging therapy) was not included as a subsequent AMT.

Patients were allowed to receive a second course of idecabtagene vicleucel (re-treatment) at the discretion of the clinical investigator, with or without bridging therapy (which had to be completed at least 14 days before the start of LDC), and if they fulfilled the following criteria:

• At least 8 weeks since first idecabtagene vicleucel infusion

• Best response to initial idecabtagene vicleucel infusion was stable disease or better, based on standard response criteria according to the IMWG uniform response criteria for MM

• Evidence of disease progression according to IMWG criteria

• No history of grade 4 CRS or neurotoxicity with prior idecabtagene vicleucel treatment

• Eligibility criteria for enrolment continued to be met (except for patients with known CNS involvement with myeloma, inadequate bone marrow function, and those who received treatment with any gene therapy-based therapeutic for cancer or investigational cellular therapy or BCMA-targeted therapy)

• Eligibility criteria for starting LDC continued to be met

• Patients with progression of myeloma within the CNS who required whole-brain or directed cerebral radiotherapy (excluding palliative focal, minimally penetrating, radiotherapy to scalp or skull lesions) were not to receive idecabtagene vicleucel re-treatment infusion until a gap of at least 8 weeks from last radiotherapy treatment had been observed.

Re-treatment required the availability of enough cryopreserved idecabtagene vicleucel drug product for re-treatment, as remanufacture of idecabtagene vicleucel from cryopreserved peripheral blood mononuclear cells and repeat leukapheresis were not allowed. Re-treatment included a second course of LDC with or without bridging therapy.

Outcomes

The objective of the KarMMa study was to evaluate the effects of idecabtagene vicleucel on response and remission rates, survival, HRQoL, and safety in patients with RRMM. The primary objective was based on the ORR. The key secondary outcome was the CRR. Other outcomes of interest to this CADTH review encompassed in the secondary objectives of the study included DoR, TTR, PFS, TTP, PFS, OS, an MRD-negative status, and HRQoL. All these outcomes were evaluated up to a cut-off date of October 16, 2019, and all efficacy analyses were conducted using the population treated with idecabtagene vicleucel. Patients were evaluated for efficacy and harm outcomes monthly until month 6, then every 3 months for a minimum of 24 months post–idecabtagene vicleucel infusion or until PD, whichever was longer. A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trial is provided in Table 8 and summarized in the following section. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

Response and Remission Rates

Response or remission rates identified as relevant by CADTH to this review were evaluated with the following end points³:

• Overall response rate, defined as the percentage of patients who achieved a PR or better. To select the best response, the following order of response was used: stringent complete response (sCR) > CR > VGPR > PR > minimal response > stable disease > PD. The outcome was assessed by an IRC according to IMWG uniform response criteria for MM. The concordance and discordance between an IRC and investigator assessment were evaluated descriptively for subjects in each category of best overall response, and separately for responders with PR or better, VGPR or better, and a CR or better. Patients were evaluated for efficacy and harm outcomes monthly until month 6, then every 3 months for a minimum of 24 months post–idecabtagene vicleucel infusion or until PD, whichever was longer.

• Complete response rate, defined as the percentage of patients who achieved a CR or sCR as assessed by an IRC according to IMWG uniform response criteria for MM. Patients were evaluated for efficacy and harm outcomes monthly until month 6, then every 3 months for a minimum of 24 months post–idecabtagene vicleucel infusion or until PD, whichever was longer.

• Time to response, defined as the time from first idecabtagene vicleucel infusion to first documentation of response of a PR or better.

• Duration of response, defined as the time from first documentation of response of PR or better to first documentation of disease progression or death from any cause, whichever occurred first. Censoring for DoR followed the censoring rules based on FDA²³ and EMA²⁴ guidance for cancer trials end points.

• Time to progression, defined as the time from first idecabtagene vicleucel infusion to first documentation of PD. Censoring rules for TTP were similar to those for PFS, except that death was not considered as an event and was censored at the last response assessment date.

• An MRD-negative status as revealed by next-generation sequencing, defined as the proportion of patients who achieved a CR or better and MRD-negative status at any time point within 3 months before achieving a CR or better until the time of PD or death (exclusive) (also referred to as MRD-negative status and ≥ CR), using a sensitivity of 10^-5, based on the population treated with idecabtagene vicleucel.

Survival

Survival outcomes relevant to this review and assessed in the KarMMa study were:

• Overall survival, defined as the time from first idecabtagene vicleucel infusion to time of death due to any cause. Patients who died on or before the data cut-off date were considered as having events on the date of death. Those who were alive or lost to follow-up were censored on the last date known alive. Those whose last date of follow-up confirmed they were alive at the data cut-off date were censored at the data cut-off date. If a patient was lost to follow-up and there was no contact after idecabtagene vicleucel infusion, the patient was censored at the idecabtagene vicleucel infusion date.

• Progression-free survival, defined as the time from first idecabtagene vicleucel infusion to first documentation of PD or death due to any cause, whichever occurred first. Guidance from the FDA²³ and EMA²⁴ for cancer trials end points was applied, and PFS for the enrolled population was also examined.

Health-Related Quality of Life

Health-related quality of life is a patient-reported outcome (PRO) that was evaluated in the PRO population using recognized and validated tools that measure HRQoL in the oncology setting: the EORTC QLQ-C30, EORTC QLQ-MY20, and EQ-5D-5L scales.

• The EORTC QLQ-C30 was measured in 121 patients from the KarMMa study. It is a generic cancer measure of HRQoL composed of both multi-item scales and single-item measures. The tool includes 5 functional scales (physical, role, emotional, cognitive, and social), 3 symptom scales (fatigue, nausea/vomiting, and pain), a global health/QoL scale, and 6 single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). Each of the multi-item scales includes a different set of items and no item occurs in more than 1 scale. The EORTC QLQ-C30 employs week recall period of 1 week for all items and a 4-point scale for the functional and symptom scales and items with response categories of “not at all,” “a little,” “quite a bit,” and “very much.” The 2 items assessing global health/QoL utilize a 7-point scale ranging from 1 (“very poor”) to 7 (“excellent”),²⁵ with higher scores on the functional and global scale indicating a higher response level. A higher score represents a high or healthy level of functioning and better HRQoL, but a high score for a symptom scale represents higher symptoms (worsening). The tool has demonstrated high reliability (internal consistency) with Cronbach alpha values greater than 0.80 for all 3 multi-item scales (ranging from 0.82 to 0.93). All 4 scales (including the single-item body image measure) had a test-retest reliability of 0.85 or greater.²⁶ Construct and criterion validity of the EORTC QLQ-C30 are adequate, as well as a good responsiveness.^27-30 Based on a mixed-methods study³¹ using a systematic review of the literature with clinician expert input, the MIDs for improvement and deterioration, respectively, were physical function (2 and −5), role function (6 and −7), cognitive function (3 and −1), emotional (6 and −3), social (3 and −6), fatigue (−4 and 5), pain (−5 and 3), nausea/vomiting (−3 and 5), financial (−3 and 2), and global QoL score (5 and −5). No studies assessing validity of the EORTC QLQ-C30 were found in patients from North America, which could affect the generalizability of the findings.

• The EORTC QLQ-MY20 was evaluated in 120 patients in the KarMMa study. It includes a myeloma module to be administered alongside the core EORTC QLQ-C30. The EORTC QLQ-MY20 is a 20-item module intended for use among patients varying in disease stage and treatment modality. It is also 1 of the most widely used measures of HRQoL in patients with MM and in clinical trials of MM to assess treatment effect on HRQoL in conjunction with the EORTC QLQ-C30. The module included in this study is composed of 20 questions addressing 4 different domains of HRQoL that are important to patients with MM: 2 symptom scales (disease symptoms and side effects of treatment) and 2 functional scales (future perspective and body image). For the symptoms scales of the EORTC QLQ-MY20 (disease symptoms and side effects of treatment), a higher score indicates worsening of symptoms (higher symptom burden), while for the functional scales (future perspectives and body image), a higher score indicates a better level of functioning. The tool extension has appropriate internal consistency³⁰ (Cronbach alpha values range from 0.77 to 0.80), validity, and responsiveness.³² The MIDs for the EORTC QLQ-MY20 were as follows: disease symptoms, −10 and + 10 points for improvement and deterioration, respectively; side effects of treatment, −10 and + 10 points for improvement and deterioration, respectively; body image, + 13 points and −13 for improvement and worsening, respectively; and future perspective (body image), + 9 and −9 points for improvement and deterioration, respectively.³³

• The EQ-5D-5L scale was evaluated in 120 patients. It is a generic preference-based HRQoL instrument that has been applied to a wide range of health conditions and treatments, including MM. A scoring function can be used to assign a value (index score) to self-reported health states from a set of population-based preference weights. The EuroQol Visual Analogue Scale (EQ VAS) has end points of 0 and 100, with respective anchors of “worst imaginable health state” and “best imaginable health state,” respectively. Respondents are asked to rate their own health by drawing a line from an anchor box to the point on the EQ VAS that best represents their own health on that day. Higher scores on the index and EQ VAS indicate better HRQoL. An MID of 0.08 for improvement and −0.10 for deterioration was defined by the sponsor and used in the KarMMa study based on recommendations from the published literature.³¹

Safety and Harms

Adverse events, SAEs, and withdrawals due to AEs were evaluated in the population treated with idecabtagene vicleucel, which included 128 patients who received idecabtagene vicleucel infusion. Data for all 140 patients who underwent leukapheresis (i.e., the enrolled population) were also included in select analyses. The report on the AEs considers those occurring in the period from LDC to immediately before the idecabtagene vicleucel infusion, which are reported as AEs related to LDC.

Health care utilization was considered in this review as a safety or harm outcome and was evaluated in the KarMMa study as an exploratory end point defined as the number of inpatient ICU days and standard care unit days.

Adverse events of special interest considered by CADTH were febrile neutropenia, neurologic effects, documented infections, replication-competent lentivirus events, development of anti-CAR antibody response, and the number of patients with CRS, all of which are evaluated in the KarMMa study.

Statistical Analysis

Sample Size

The sample size was calculated for the primary efficacy end point, ORR, based on a 1-sample binomial test with normal approximation. The null hypothesis was that the ORR (the proportion of patients with a PR or better based on all idecabtagene vicleucel–treated patients) is no more than 50%; the alternative hypothesis was that the ORR is greater than 50%, with a target ORR of 70%. The choice of a null hypothesis of 50% ORR was based on the observed clinical activity of the best available single-drug therapy in a heavily pretreated RRMM patient population. The calculated sample size was 119 idecabtagene vicleucel–treated patients. This number would have provided greater than 99% power at a 1-sided 0.025 alpha level. Assuming a dropout rate of 15% between the time of study enrolment and idecabtagene vicleucel infusion, a total of 140 patients were required to be enrolled.

Investigators planned the primary analysis for when all idecabtagene vicleucel–treated patients had sufficient follow-up. For the primary analysis, a minimum follow-up of 10 months after the last patient received idecabtagene vicleucel infusion (9 months from the time of the last patient’s first anticipated response) was considered sufficient. A data cut-off date of October 16, 2019, was planned for the primary analysis.

Statistical Tests and Analyses

Demographics, baseline characteristics, and medical history were summarized for the population treated with idecabtagene vicleucel. For the enrolled population, only demographics and baseline characteristics are presented. Continuous demographic and disease characteristics variables (e.g., age, time from diagnosis and number of prior regimens) at baseline were summarized using descriptive statistics, while categorical variables (e.g., age categories [< 65 and ≥ 65; < 75 and ≥ 75], sex, race, ECOG score, and Revised International Staging System stage) were summarized by frequency counts and percentages.

All efficacy analyses were conducted using the population treated with idecabtagene vicleucel. Supportive analyses of the primary (ORR), key secondary (CRR), and other secondary (PFS and OS) end points using the enrolled population were also conducted. Duration of response and TTR were measured on responders (i.e., patients who achieved a PR or better) and therefore were the same for the population treated with idecabtagene vicleucel, enrolled population, and efficacy evaluable population. The primary efficacy analysis was based on the IRC-adjudicated assessment according to IMWG uniform response criteria for MM in the population treated with idecabtagene vicleucel. There were no adjustments for multiple comparisons or procedure to control the type I error rate for the secondary outcomes beyond the assessment of the CR.

The end points were assessed as follows:

• The ORR was calculated as the number of patients who achieved PR or better divided by the number of patients in the specific analysis population, using a 1-sample binomial test. The ORR was first tested against the null hypothesis of an ORR of no more than 50% at a 1-sided alpha level of 0.025. If the ORR test was significant, the CRR was to be subsequently tested against the null hypothesis of a CRR of no more than 10% at the same 1-sided alpha level of 0.025. The 2-sided Wald 95% CIs for all ORR values were calculated.

• The CRR was tested against the null hypothesis of no more than 10% for the key secondary efficacy end point. It was calculated as the number of patients who achieved a CR or sCR divided by the number of patients in the specific analysis population. The CRR based on the IMWG criteria (i.e., CR and sCR) was calculated together with the 2-sided 95% CI. If the test for the ORR (primary end point) was positive, the CRR was to be tested against the null hypothesis of no more than 10% with a 1-sample binomial test using a stepdown approach to control the overall alpha level, which would remain at the 1-sided 0.025 level.

• The DoR was analyzed (for responders only) using the KM method. Median DoR and the corresponding 95% CIs are provided. Censoring for DoR followed the censoring rules based on FDA²³ and EMA²⁴ guidance for cancer trials end points.

• The TTR (assessed in responders only) was summarized using descriptive statistics.

• The TTP censoring rules were similar to those for DoR, except that death was not considered an event and was censored at the last response assessment date. The TTP was summarized for the population treated with idecabtagene vicleucel using KM statistics; the KM approach was also applied to the median and 95% CI.

• Progression-free survival was summarized for the population treated with idecabtagene vicleucel using KM statistics and a KM curve was used to depict the outcome. The median PFS along with the 2-sided 95% CI for the median was estimated. Censoring rules were likewise based on FDA²³ and EMA²⁴ guidance for cancer trials end points. Data for PFS in the enrolled population were also examined. For this analysis, the PFS was calculated from the date of enrolment (i.e., leukapheresis).

• The MRD-negative rate was calculated and presented with the 95% CI. A 1-sample exact binomial test was performed for the MRD-negative rate as a descriptive analysis, with a null hypothesis of no more than 10% and an alternative hypothesis of greater than 10%.

• Overall survival was analyzed for the population treated with idecabtagene vicleucel using the KM method. Median OS and the corresponding 95% CI are provided. Survival rates at specific time points and KM survival curves are also provided. Overall survival for the enrolled population was analyzed. For this analysis, the OS was calculated from the date of enrolment (i.e., leukapheresis). The censoring rules remained the same as for the analysis in the population treated with idecabtagene vicleucel. If the patient was lost to follow-up and there was no contact after enrolment, the patient was censored at the date of leukapheresis instead of the date of idecabtagene vicleucel infusion.

• The PROs using HRQoL measures were evaluated with the EORTC QLQ-C30, EORTC QLQ-MY20, and EQ-5D-5L scales. The PRO end points of primary interest focused on the EORTC QLQ-C30 subscales of fatigue, pain, physical functioning, cognitive functioning, and global health/QoL. The EORTC QLQ-MY20 used the symptom subscales of disease symptoms and side effects. The EQ-5D-5L tool used the index and EQ VAS to depict the results. The analysis was descriptive and performed primarily to assess the total and subdomain scores on the primary PRO end points at each time point, describe the changes from baseline to each subsequent time point for the total and subdomain scores on the primary PRO end points, and evaluate the proportion of patients reaching the responder definitions in the changes from baseline to each subsequent time point for the total and subdomain scores on the primary PRO end points.

Subgroup Analysis

Prespecified subgroup analysis for the primary outcome were performed for subgroups in which an adequate number of patients were available to allow for meaningful interpretation of results. Analyses were performed for the following subgroups of interest to this review within the population treated with idecabtagene vicleucel:

• age group 1: younger than 65 years, at least 65 years old; and age group 2: younger than 65 years, 65 to 69 years, 70 years and older

• extramedullary plasmacytoma (yes, no)

• number of prior AMT regimens per year (≤ 1, > 1)

• revised International Staging System stage at baseline (I or II, III)

• tumour BCMA expression (≥ 50%, < 50% BCMA+).

The ORR, CRR, and DoR were evaluated for each of the subgroups. Forest plots and 95% CIs were provided for the subgroup analyses of the ORR and CRR. Additional subgroup analyses could be explored as appropriate, but no other subgroups were analyzed a priori besides those already stated. No statistical test was performed to evaluate differences in effect estimates on these outcomes between subgroups.

Sensitivity Analyses

Alternate analyses were run for the outcome of PFS where the application of different guidance rules for censoring were applied using FDA and EMA rules.

Analysis Populations

Analysis populations included the enrolled population (all patients in the screened population who underwent leukapheresis), the population treated with idecabtagene vicleucel (all patients in the enrolled population who received idecabtagene vicleucel infusion), the population re-treated with idecabtagene vicleucel, (all patients who received idecabtagene vicleucel re-treatment), and the PRO analysis population (those who completed their baseline PRO questionnaires using HRQoL measures). Efficacy and safety analyses were evaluated in the population treated with idecabtagene vicleucel and in each idecabtagene vicleucel target dose group (i.e., 150 × 10⁶, 300 × 10⁶, and 450 × 10⁶ CAR T cells).

Results

Patient Disposition

The KarMMa study screened 158 patients, of whom 18 failed to meet the eligibility criteria, as shown in Figure 3.

A total of 140 patients underwent leukapheresis and were included in the enrolled population. Of these, 128 patients received idecabtagene vicleucel infusion and were included in the idecabtagene vicleucel–treated population used for the primary analysis of efficacy and safety (Table 9).

Reasons for discontinuation for the pretreatment period (from leukapheresis to before LDC) and treatment period (from LDC to idecabtagene vicleucel infusion) for the enrolled population, and the posttreatment follow-up period for the population treated with idecabtagene vicleucel after initial idecabtagene vicleucel infusion are described. For re-treatment patients, discontinuation reasons are summarized separately.

Figure 3: Patient Disposition of Study BB2121-MM-001 (KarMMa)

AE = adverse event, Ide-cel = idecabtagene vicleucel; LD = lymphodepleting; LTFU = lost to follow-up.

Note: Data cut-off date = October 16, 2019.

* Includes 41 patients who discontinued after initial idecabtagene vicleucel infusion without entering the re-treatment period and 17 patients who discontinued during the re-treatment period. The re-treatment period includes re-treatment screening, re-treatment baseline, re-treatment, and re-treatment follow-up.

** Includes 58 patients ongoing after initial idecabtagene vicleucel infusion without entering the re-treatment period and 12 patients ongoing after entering the re-treatment period. Patients ongoing in re-treatment follow-up include patients who were screened for re-treatment and have not discontinued the study as of the data cut-off date, regardless of whether or not the patient actually received idecabtagene vicleucel re-treatment.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

The majority of patients (91.4%) who underwent leukapheresis received the idecabtagene vicleucel infusion. Of the 12 patients who discontinued after leukapheresis, 8 did so before LDC. As of the October 16, 2019, cut-off date, 70 of the 128 idecabtagene vicleucel–treated patients (54.7%) were still participating in the study, including 58 (45.3%) ongoing after initial idecabtagene vicleucel infusion without entering the re-treatment period and 12 (9.4%) ongoing after entering the re-treatment period. Fifty-eight (45.3%) of the 128 idecabtagene vicleucel–treated patients discontinued the study, with 31 patients (24.2%) discontinuing due to death, 1 patient (0.8%) lost to follow-up, and 26 patients (20.3%) discontinuing due to withdrawal by subject. Although there were 26 withdrawals by patient, only 2 were censored for PFS due to discontinuing study without PD or death. The median time from leukapheresis to idecabtagene vicleucel product release was 32.0 days, from product release to idecabtagene vicleucel infusion the median time was 7.0 days, and from leukapheresis to idecabtagene vicleucel infusion it was 40.0 days.

The disposition of patients who were re-treated after a first idecabtagene vicleucel administration is shown in Table 10.

Exposure to Study Treatments

Lymphodepleting Chemotherapy

Of the 140 patients who were included and underwent leukapheresis, 132 received LDC, and of these 128 went on to receive the idecabtagene vicleucel infusion. Patient exposure to cyclophosphamide and fludarabine was consistent with the lymphodepletion treatment plan according to the protocol, and the number of patients and number of days dosed are supplied in Table 11.

Idecabtagene Vicleucel Infusion

Idecabtagene vicleucel was administered at planned target doses of 150 × 10⁶, 300 × 10⁶, or 450 × 10⁶ CAR T cells per infusion. The median time from leukapheresis to idecabtagene vicleucel administration in the idecabtagene vicleucel–treated population was 40.0 days (range = 33 to 79). Seven of 128 patients (5.5%) were administered idecabtagene vicleucel more than 8 weeks (56 days) after leukapheresis and 12 of 128 patients (9.4%) were administered idecabtagene vicleucel within 5 weeks (35 days) of leukapheresis. One of the idecabtagene vicleucel–treated patients received a second leukapheresis due to manufacturing failure after the first leukapheresis. One patient did not receive idecabtagene vicleucel due to manufacturing failure.

Of the 128 patients who received idecabtagene vicleucel infusion, 4 (3.1%) had a target dose of 150 × 10⁶ CAR T cells, 70 (54.7%) had a target dose of 300 × 10⁶ CAR T cells, and 54 (42.2%) had a target dose of 450 × 10⁶ CAR T cells. All idecabtagene vicleucel–treated patients received an actual dose between 150.5 × 10⁶ and 518.4 × 10⁶ CAR T cells, which was within the allowance of 20% over the target dose of 450 × 10⁶ CAR T cells (i.e., less than 540 × 10⁶ CAR T cells). The median actual dose received across all target dose levels was 315.3 × 10⁶ CAR T cells. A total of 51 patients received an actual dose in excess of 450 × 10⁶ CAR T cells; 4 patients received an actual dose of more than 500 × 10⁶ CAR T cells. No patients had a dose interruption during idecabtagene vicleucel infusion.

Bridging Therapy

Bridging therapy was considered for a patient, given their expected rate of deterioration, while idecabtagene vicleucel was being manufactured, based on the clinical judgment of the respective investigator. Bridging therapies could have included corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies as single agents or in combination. The majority of patients (87.5%) received bridging therapy for myeloma control during the idecabtagene vicleucel manufacturing period. The most commonly received bridging therapies were dexamethasone (70.3%), cyclophosphamide (36.7%), daratumumab (28.1%), carfilzomib (23.4%), bortezomib (19.5%), and pomalidomide (18.8%). The median duration of bridging therapy was 14.5 days (range = 1 to 33).

Subsequent Antimyeloma Therapies

The most commonly received (in ≥ 5% of patients) subsequent AMTs are shown in Table 12. Idecabtagene vicleucel re-treatment is not included as a subsequent therapy.

When considering the time to subsequent AMT in the population treated with idecabtagene vicleucel, idecabtagene vicleucel re-treatment (as well as bridging therapy for re-treatment and LDC before re-treatment) was counted as a subsequent AMT. As of the data cut-off date, 56 patients (43.8%) received a subsequent AMT (including idecabtagene vicleucel re-treatment, bridging AMT for re-treatment, or LDC before re-treatment). The KM estimate for median time to subsequent AMT was 13.2 months (95% CI, 10.9 to 15.1).

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported below.

Data for cut-off dates of January 14, 2020, and April 7, 2020. are presented in Appendix 3.

Response and Remission Rates

Overall Response Rate

The primary efficacy end point was the ORR based on the IRC-adjudicated assessment of response. The ORR in the idecabtagene vicleucel–treated population was 73.4% (95% CI, 65.8 to 81.1; P < 0.0001). An ORR of 50.0% was observed at the target dose of 150 × 10⁶ CAR T cells, 68.6% was observed at the target dose of 300 × 10⁶ CAR T cells, and 81.5% was observed at the target dose of 450 × 10⁶ CAR T cells. The results of this primary analysis in the population treated with idecabtagene vicleucel were consistent with those of the supportive analyses in the enrolled population. A total of 66 patients (51.6%; 95% CI, 42.9 to 60.2) achieved a VGPR or better.

A forest plot for ORRs for the idecabtagene vicleucel–treated population based on pre-planned subgroups, including those stated in the protocol for this review, is presented in Figure 22 in Appendix 3. The treatment effect for the ORR was generally consistent across subgroups, but appeared to be heterogeneous according to stage at diagnosis, based on the International Staging System stage. However, there is only a small number of patients in the stage III group and no a priori statistical tests were planned to provide definitive conclusions about any differences in subgroup effects.

Complete Response Rate

In the population treated with idecabtagene vicleucel, 40 patients out of 128 achieved a CR, equivalent to a CRR of 31.3% (95% CI, 23.2 to 39.3; P < 0.0001). A CR rate of 25.0% was observed at the target dose of 150 × 10⁶ CAR T cells, 28.6% was observed at the target dose of 300 × 10⁶ CAR T cells, and 35.2% was observed at the target dose of 450 × 10⁶ CAR T cells. The results of this primary analysis of the CRR in the idecabtagene vicleucel–treated population were consistent with those of the supportive analyses in the enrolled population.

Duration of Response

The DoR analysis was based on responders only, and therefore was the same for the idecabtagene vicleucel–treated and enrolled population. A summary of DoR among responders based on an IRC assessment, applying FDA censoring rules, is presented in Table 13 and Table 14. As of the data cut-off date, 52 (55.3%) of 94 responders had progressed or died and 42 (44.7%) were censored. The KM estimate for median DoR among responders was 10.6 months (95% CI, 9.0 to 11.3). Based on KM estimates, ||||| of responders had responses lasting for at least 9 months, and ||||| of responders had responses lasting for 12 months or longer. The median DoR among responders was 9.9 months (95% CI, 5.4 to 11.0) at the target dose of 300 × 10⁶ CAR T cells and 11.3 months (95% CI, 9.2 to 11.4) at the target dose of 450 × 10⁶ CAR T cells. The median was NE at the target dose of 150 × 10⁶ CAR T cells. A majority (59.1%) of responders at the target dose of 450 × 10⁶ CAR T cells were still progression-free as of the data cut-off date.

Applying EMA censoring rules produced similar results when compared to applying the FDA rules. As of the data cut-off date, 54 (57.4%) of 94 responders had progressed or died and 40 (42.6%) were censored. The KM estimate for median DoR among responders was 10.5 months (95% CI, 8.0 to 11.3). Based on KM estimates, 60.5% of responders had responses lasting for at least 9 months, and 32.0% of responders had responses lasting for 12 months or longer.

Time to Response

The TTR was calculated as the time from idecabtagene vicleucel infusion to the first date of response (PR or better) and this analysis was based on responders only. A summary of TTR among responders based on the IRC assessment is presented in Table 13. As of the data cut-off date, the median TTR among patients who responded was 1.0 months (range = 0.5 to 8.8), which was the time of the first scheduled response assessment. Within 1 month after idecabtagene vicleucel infusion, 84.0% of responders achieved a first response. An additional |||||| of responders achieved a first response between 1 and 2 months after idecabtagene vicleucel infusion. The median TTR was 1.0 months for responders in each of the target dose groups.

Time to Progression

The TTP was defined as time from idecabtagene vicleucel infusion to the first documented progression. A summary of TTPs among responders based on the IRC assessment, applying FDA censoring rules, is presented in Table 13. The KM estimate for median TTP was 8.9 months (95% CI, 6.0 to 11.6). Based on KM estimates, 49.1% of patients remained event-free at 9 months and 35.5% of patients remained event-free at 12 months. Results produced by applying the EMA censoring rules were similar to those produced by applying FDA censoring rules; supportive analyses were also similar to primary analyses for this outcome.

Minimal Residual Disease–Negative Status

The primary analysis of MRD status was the proportion of patients who achieved a CR or better and MRD-negative status at any time point within 3 months before achieving a CR or better until the time of PD or death (referred to as MRD-negative status and a CR or better), based on the idecabtagene vicleucel–treated population and measurements using ne|t-generation sequencing with a sensitivity level of 10^-5 nucleated cells. A summary of the primary analysis of MRD is presented in Table 13. As of the data cut-off date, 40 of 128 patients (31.3%) achieved a CR or better; 31 of 128 patients (24.2%) achieved an MRD-negative status and a CR or better (95% CI, 17.1 to 32.6), rejecting the null hypothesis of no more than 10% for this end point (P < 0.0001). The proportion of patients who achieved an MRD-negative status and a CR or better was 1 of 4 patients (25.0%) at the target dose of 150 × 10⁶ CAR T cells, 17 of 70 patients (24.3%) at the target dose of 300 × 10⁶ CAR T cells, and 13 of 54 patients (24.1%) at the target dose of 450 × 10⁶ CAR T cells.

Figure 4: [Redacted]

This figure has been redacted as per the sponsor’s request

Survival

Overall Survival

Overall survival for the idecabtagene vicleucel–treated population was defined as the time from idecabtagene vicleucel infusion to death due to any cause. For OS in the enrolled population, the OS was calculated from the date of enrolment (i.e., leukapheresis). A summary of OS in the population treated with idecabtagene vicleucel is presented in Table 14 and displayed in Figure 5. As of the data cut-off date, 34 patients (26.6%) had died. The KM estimate for the median OS was 18.2 months (95% CI, 18.0 to NE), with a median follow-up time for all surviving patients of 12.0 months. Based on KM estimates, 83.9% of patients were alive at 9 months and 76.9% of patients were alive at 12 months; 71.6% and 54.5% of patients were alive at 15 and 18 months, respectively. The median OS was 18.2 months (95% CI, 9.4 to 18.2) at the target dose of 150 × 10⁶ CAR T cells. The median OS was NE (95% CI, 16.8 to NE) at the target dose of 300 × 10⁶ CAR T cells, or at the target dose of 450 × 10⁶ CAR T cells. Based on KM estimates, 78.6% and 76.8% of patients were alive at 12 months in the 300 × 10⁶ and 450 × 10⁶ CAR T cells target dose groups, respectively. In the enrolled population, as of the data cut-off date, 42 patients (30.0%) had died and the KM estimate for median OS was 19.3 months (95% CI, 17.94 to NE). Based on KM estimates, 81.3% of patients were alive at 9 months and 75.5% of patients were alive at 12 months; 67.3% and 61.8% of patients were alive at 15 months and 18 months, respectively.

Progression-Free Survival

Progression-free survival was defined as the time from idecabtagene vicleucel infusion to the first date of documented PD or death from any cause during the study, whichever occurred earlier. A summary of the PFS analysis in the population treated with idecabtagene vicleucel based on an IRC assessment, applying FDA censoring rules, is presented in Table 14 and displayed in Figure 6. As of the data cut-off date with a median PFS follow-up duration of 10.2 months (range = 0.03 to 18.3) for censored patients, 83 (64.8%) had a PFS event and 45 (35.2%) were censored. The KM estimate for median PFS was 8.6 months (95% CI, 5.6 to 11.3). Based on KM estimates, 47.4% of patients remained event-free at 9 months and 34.3% of patients remained event-free at 12 months. The median PFS was 5.8 months (95% CI, 4.2 to 8.9) at the target dose of 300 × 10⁶ CAR T cells and 11.3 months (95% CI, 8.8 to 12.4), with 53.7% of patients progression-free at the target dose of 450 × 10⁶ CAR T cells. The median PFS was 2.8 months (95% CI, 1.0 to NE) at the target dose of 150 × 10⁶ CAR T cells, and 11.3 months (95% CI, 8.8, 12.4), with a slight majority of patients (53.7%) still progression-free, at the target dose of 450 × 10⁶ CAR T cells. The most common reason for censoring was that patients were ongoing at the data cut-off date (39 patients; 30.5%). Two patients (1.6%) were censored due to discontinuing study without an event, 2 patients (1.6%) were censored due to taking anti-MM therapy before progression or death, and 2 patients (1.6%) were censored due to missing assessments. Analysis of the PFS using the EMA censoring rules produced results similar to those of the analysis using FDA censoring rules, with a median using the KM estimate of 8.2 months (95% CI 5.6 to 11.0).

Figure 5: [Redacted]

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

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

This figure was redacted as per the sponsor request.

Figure 6: [Redacted]

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

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

This figure was redacted as per the sponsor’s request

In an updated cut-off data submission (January 14, 2020)¹⁰ with a median follow-up of 13.3 months, no significant changes were noted in the ORR, CRR, DoR, TTR, MRD, OS, or PFS. These updated results are described in Appendix 3. As stated in the KarMMa 001 protocol, these data were not adjusted for alpha errors nor used for hypothesis testing. Also, data for cut-off dates of April 7, 2020, presented in Appendix 3, show no significant differences from the initial cut-off date analysis.

Intravenous Immunoglobin Replacement

Intravenous immunoglobin replacement was not assessed in the KarMMa study.

Health-Related Quality of Life

Health-related quality of life was evaluated with the EORTC QLQ-C30, the EORTC QLQ-MY20, and the EQ-5D-5L. Baseline levels for all 3 scales are shown in Table 15 and the changes from baseline for each measurement are presented in Figure 7 through Figure 15. These end points were considered descriptive in nature without adjustment for multiplicity, and were not evaluated as time-to-event outcomes.

European Organisation for Research and Treatment of Cancer QLQ-C30

Overall, based on data from the available patients described for this end point, a decrease in the fatigue subscale scores was observed in reference to the MID of −4 points from baseline to month 2, reaching a nadir at months 3, 4, 5, 6, and 9, and remaining below the MID for improvement at months 12 and 15 (Figure 7).

For the pain subscale, a similar direction in the scores described in the available population was observed, with an MID of −5 points from baseline to month 15. These changes in the score began at month 1 and reached a nadir by month 3. (Figure 8).

For the physical functioning subscale, an increase in mean scores from baseline of 2 points for the MID was reported beginning at month 1. The nadir was reached by month 2 and continued until month 12 (Figure 9). Cognitive functioning subscale scores generally demonstrated no change beyond or above the MID of 3 for improvement or −1 for deterioration, from baseline to month 15 (Figure 10).

Finally, an increase in global health/QoL subscale scores relative to the MID of 5 points from improvement began at month 2, reaching a nadir at months 4, 5, 6, and 9. The mean subscale score increases remained at months 12 and 15 (Figure 11).

European Organisation for Research and Treatment of Cancer QLQ-MY20

Small decreases crossing the MID reference of −10 points were observed in the disease symptoms subscale from baseline to months 3 through 15 posttreatment (Figure 12). A gradual increase in mean side effects subscale scores was observed from baseline to months 1 through 9 posttreatment; however, these increases did not go beyond the MID reference for deterioration of 10 points (Figure 13).

EuroQol 5-Dimensions 5-Levels Questionnaire

The EQ-5D-5L was measured using an index and the EQ VAS. For the index, a slight increase in mean scores was observed from baseline throughout follow-up, reaching a level above the MID, for an improvement of 0.08 beginning at month 2 (Figure 14). Beyond month 2, the index score remained relatively stable and marginally above the MID for improvement. For the EQ VAS, an increase in the MID in mean scores from baseline of 7 points was reported beginning at month 3 and at month 5 (Figure 15). Beyond month 5, the EQ VAS scores remained above the MID

Figure 7: Mean Change in Scores From Baseline on the EORTC QLQ-C30 Fatigue Subscale by Time Point

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 8: Mean Change in Scores From Baseline on the EORTC QLQ-C30 Pain Subscale by Time Point

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 9: Mean Change in Scores From Baseline on the EORTC QLQ-C30 Physical Functioning Subscale by Time Point

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 10: Mean Change in Scores From Baseline on the EORTC QLQ-C30 Cognitive Functioning Subscale by Time Point

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer – Quality of Life C30 questionnaire; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 11: Mean Change in Scores From Baseline on the EORTC QLQ-C30 Global Health/QoL Subscale by Time Point

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LDC = lymphodepleting chemotherapy; MID = minimal important difference; QoL = quality of life.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 12: Mean Change in Scores From Baseline on the EORTC QLQ-MY20 Disease Symptoms Subscale by Time Point

EORTC QLQ-MY20 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Myeloma Module with 20 items; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 13: Mean Change in Scores From Baseline on the EORTC QLQ-MY20 Side Effects Subscale by Time Point

EORTC QLQ-MY20 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Myeloma Module with 20 items; LDC = lymphodepleting chemotherapy; MID = minimal important difference.

Note: Baseline defined as the last non-missing assessment on or before day of LDC. Error bars represent 95% confidence intervals. Data cut-off date: October 16, 2019.

Source: Clinical Study Report for BB2121-MM-001 (KarMMa).¹¹

Figure 14: [Redacted]

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

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

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

|||||||||||||||||||||¹¹

Note: This figure has been redacted as per the sponsor’s request.

Figure 15: [Redacted]

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

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

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

Note: This figure has been redacted as per the sponsor’s request.

Harms

Harms identified in the protocol of this review are reported below. In the KarMMa trial, the primary safety analysis was based on the population treated with idecabtagene vicleucel, which included 128 patients who received an infusion of idecabtagene vicleucel. The safety evaluation included all safety data as of the data cut-off date of October 16, 2019. The median duration of follow-up after idecabtagene vicleucel infusion for all idecabtagene vicleucel–treated patients was 11.3 months (range = 0.2 to 18.6), and 17.8 months, 13.9 months, and 9.7 months in patients who received target doses of 150 × 10⁶, 300 × 10⁶, and 450 × 10⁶ CAR T cells, respectively.

Table 16 provides detailed harms data. All data pertain to safety events occurring after the initial idecabtagene vicleucel administration; safety data during the re-treatment period are reported separately.

Adverse Events

All 128 patients treated with idecabtagene vicleucel reported grade 3 or 4 AEs occurring in 127 patients (99.2%). Most AEs, with the exception of hypogammaglobulinemia and infections, occurred within the first 8 weeks after infusion.

The most commonly reported AEs were hematologic toxic effects (Table 16), including neutropenia in 117 patients (91.4%), CRS in 107 patients (83.6%), anemia in 89 patients (69.5%), and thrombocytopenia in 81 patients (63.3%). Among patients with persistent grade 3 or grade 4 neutropenia (52 patients) or thrombocytopenia (62 patients) 1 month after infusion, the median time to recovery to grade 2 or lower was 1.9 months (range = 1.2 to 5.6) and 2.1 months (range = 1.2 to 13.8), respectively.

Serious Adverse Events

In the period on or after idecabtagene vicleucel infusion, 86 patients (67.2%) had at least 1 SAE. The most frequently reported SAEs (≥ 5% of patients) were CRS in 22 patients (17.2%), general physical health deterioration in 13 patients (10.2%), pneumonia in 11 (8.6%), and febrile neutropenia in 9 (7.0%). The frequencies of patients with SAEs were generally similar between the target dose subgroups, except for those who received a target dose of 450 × 10⁶ CAR T cells, who had at least 5% higher frequencies of influenza when compared to the 300 × 10⁶ CAR T cells group (5.6% versus 0% respectively), pyrexia (7.4% versus 1.4%), and basal cell carcinoma (7.4% versus 1.4%).

Mortality

Eight patients died after leukapheresis and before receiving idecabtagene vicleucel infusion: 5 patients (3.6%) died after leukapheresis and before starting LDC, and 3 patients (2.1%) died after starting LDC and before receiving idecabtagene vicleucel infusion. In 6 of these 8 patients, the death was classified as “death from malignant disease under study” or “complication due to malignant disease under study.” One of the other 2 deaths was categorized as “death from AE” due to acute respiratory failure, and the other as “death from other cause” due to an event coded as cerebral hemorrhage.

In total, in the population treated with idecabtagene vicleucel as of the data cut-off date, 34 patients (26.6%) died on or after idecabtagene vicleucel infusion, with 24 of these deaths attributed to the malignant disease under study or complications due to the malignant disease under study. The primary cause of death was attributed to “death from AE” in 6 patients (4.7%), including deaths due to bronchopulmonary aspergillosis, pneumonia cytomegaloviral, sepsis, gastrointestinal hemorrhage, CRS, and subdural hematoma (all occurring in 1 patient each, 0.8%).

Notable Harms

Harms of special interest were identified according to the protocol for this review. Febrile neutropenia was present in 21 patients (16.4%) in the population treated with idecabtagene vicleucel. For the outcome of neurotoxicity, a total of 23 patients (18.0%) in the idecabtagene vicleucel–treated population had investigator-identified neurotoxicity on or after idecabtagene vicleucel infusion. Psychiatric disorders included confusional states or anxiety (in 45 patients [35.2%]) and nervous system disorders (in 67 patients [52.3%]). The replication-competent lentivirus is a hypothetical complication occurring in patients receiving CAR T-cell therapies, but no cases were detected. Hospital utilization was measured with the number of ICU admissions, which 25 patients (19.5%) required. Infections or infestations were present in 88 patients (68.8%), most of which were upper respiratory infections (18 [14.1%]) and pneumonia (15 [11.7%]).

Cases of CRS on or after idecabtagene vicleucel infusion were reported in 107 patients (83.6%). The frequencies of patients with CRS-related AEs increased by target dose level: 50.0%, 75.7%, and 96.3% at the target doses of 150 × 10⁶, 300 × 10⁶, and 450 × 10⁶ CAR T cells, respectively. The most frequent treatment for patients with CRS was tocilizumab (52%), followed by corticosteroids (14.8%), anakinra (1.6%), and siltuximab (0.8%). The majority of patients had CRS of grade 1 or 2 maximum severity (100 patients; 78.1%), while grade 3 or 4 maximum severity CRS was reported in 6 patients (4.7%). All CRS-related AEs occurred within the first 8 weeks after idecabtagene vicleucel infusion, with a median duration of all CRS cases of 5 days (range = 1 to 63 days).

In an updated data submission (cut-off date: January 14, 2020),¹⁰ no differences in the number of patients with AEs were noted. Ten more deaths were reported at this cut-off date (44 [34.4%]) and 23 more cases of febrile neutropenia (a total of 30 of 128 treated patients [16.3%]) were reported. The other harms of special interest were not significantly different at this cut-off date. These results are described in Appendix 3.

Critical Appraisal

Internal Validity

The KarMMa study was a single-arm, open-label, phase II study. Limitations of the study stem from the single-arm design and lack of a comparator group, which reduce the confidence in the effect estimates due to an increase in the risk of bias associated with confounding fluctuations in health status, and unidentified prognostic factors, which could affect subjectively assessed outcomes. The open-label design may introduce bias and subjectivity when assessing HRQoL outcomes as it was not evaluated in a blinded fashion. The HRQoL outcomes were considered secondary end points and analyzed in smaller groups (N = 121). The use of well-validated measurement scales for assessing PROs were appropriate and may mitigate bias in the measurement of these outcomes.

Patients considered in the enrolled population (N = 140) and those included in the analyzed idecabtagene vicleucel–treated population (N = 128) were similar across different measures and outcomes, implying a low risk of bias due to differences in patients who withdrew from the study after enrolment. All patients in the population treated with idecabtagene vicleucel were analyzed, also suggesting a low risk of attrition bias and selection of participants into the cohort of treated patients. However, the number of patients included in the different target doses groups were dissimilar, mainly due to changes during the amendments in the study (Amendment 2.0), with small sample sizes in these individual groups (e.g., only 4 patients included in the 150 × 10⁶ CAR T cells target dose group) which creates issues of imprecision in effect estimates in these groups. The patients from the 450 × 10⁶ CAR T cells target dose group (n = 54) was similar overall in baseline and demographic characteristics. Similarly, the PROs (HRQoL outcomes) were not considered in a multiplicity adjustment, had significant amounts of missing data at baseline and a significant proportion missing at a later time, creating a potential for bias in those that remain and who may not be reflective of the overall population (i.e., those that died tend to have had a poorer HRQoL).

Investigators tested their primary efficacy end point, the ORR, against the null hypothesis of an ORR of less than 50% (i.e., a 50% improvement over daratumumab monotherapy) and if this was significant, the CRR would be tested against a null hypothesis of a CRR of less than 10% (a 10% improvement of idecabtagene vicleucel over daratumumab monotherapy). These historical values for the null hypothesis appear to be appropriate as they were obtained from recent clinical trials.

Bridging therapy for myeloma control following leukapheresis while idecabtagene vicleucel was being manufactured was used in the majority of patients (87.5%), and these included corticosteroids, alkylating agents, IMiDs, proteasome inhibitors, and/or anti-CD38 antibodies. Whether these interventions could have an effect on the outcomes evaluated in these patients after receiving idecabtagene vicleucel treatment is not known.

The prospective nature of the study defines the time of assigning and starting the intervention and the follow-up, which ameliorates any misclassification of variables and other measurements of different outcomes. Few deviations from the intended interventions were detected and all appeared to be properly reported. Deviations from the protocol occurred in 46 patients (35.9%) of the population treated with idecabtagene vicleucel and most were due to reporting errors. There were instances of missing data in the HRQoL, but there was adequate reporting of the censoring rules throughout the study according to FDA censoring rules for time-to-event outcomes (DoR, OS, and PFS).

Multiplicity was addressed only for the ORR and CR end points, using a stepdown procedure to control the family-wise type I error rate, testing first the primary end point of ORR, with the key secondary end point tested only if the test of primary efficacy end point was positive. The lack of multiplicity adjustments, (i.e., the type I error rate) for the other secondary and exploratory outcomes relevant to this report (DoR, TTR, TTP, MRD status, OS, PFS, and HRQoL) generates uncertainty due to the increased possibility of observing a chance association as causal.

Subgroups were planned a priori, but no formal subgroup effect tests were conducted due to the expected small number of patients and events (imprecision). Some subgroups were relatively small, producing relatively wide and imprecise CIs around the point estimates. Given the relatively small numbers of patients, it is unclear if they are representative of the larger population.

External Validity

According to input from the clinical experts, patients in the KarMMa trial were generally representative of those who are anticipated to receive idecabtagene vicleucel in the Canadian landscape (i.e., the population with RRMM of Canada) who have already received multiple lines of treatment but are usually considered in a health state to be able to receive the CAR T-cell therapy. Overall, the clinical variables and patient baseline characteristics from the KarMMa trial were similar to what would be expected to occur in clinical practice in patients who are already refractory to IMiD, proteasome inhibitors, and anti-CD38 drugs, including the performance status (Table 17).

The administration of idecabtagene vicleucel in the KarMMa study involves several additional actions to consider. First, the manufacturing process takes 4 to 5 weeks to complete, from the leukapheresis to the idecabtagene vicleucel infusion. Second, the use of bridging therapies requires treatments that are currently available in the Canadian clinical practice (i.e., corticosteroids, IMiD, proteasome inhibitors, and anti-CD38 antibodies). Third, LDC should be used in all patients before receiving idecabtagene vicleucel. These actions present further variables to consider in the feasibility and acceptability of the intervention as they would be a constant part of the new treatment. According to feedback from the clinical experts consulted by CADTH, these steps are necessary components of the idecabtagene vicleucel administration process and would be implemented in Canadian clinical practice.

Re-treatment of patients was allowed in the KarMMa study if they fulfilled specific clinical criteria, which included having enough cryopreserved material to produce another dose of idecabtagene vicleucel. This strategy is likely to be considered by physicians and patients receiving idecabtagene vicleucel. Complying with the eligibility criteria for re-treatment will be important for the applicability of this option.

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

This section summarizes and appraises indirect evidence comparing the relative efficacy and safety of idecabtagene vicleucel against relevant comparators.

One of the limitations of the KarMMa study is the lack of a comparator arm to evaluate the relative efficacy of idecabtagene vicleucel in patients with RRMM to standards of care.

Two sponsor-submitted analyses are included in this report. The first (NDS-MM-003)¹² aims to evaluate the effect of idecabtagene vicleucel from the KarMMa study (MM-001), which has a single-arm design, and compare its results with the effectiveness of other currently administered treatments, using patient-level RWE collected from various sources in a comparable population (RRMM according to the inclusion and exclusion criteria of the KarMMa study). The second is a sponsor-submitted MAIC, which is described and evaluated in this section as a high-level critical appraisal.¹³

A supplemental literature search was conducted by CADTH to identify further ITCs of idecabtagene vicleucel in RRMM patients. A focused literature search for ITCs dealing with MM was run in MEDLINE All (1946–) on January 21, 2021. The search was limited to documents published between January 1, 2011, and January 21, 2021. Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer will be acquired. Reviewers made the final selection of studies to be included in the review, and differences were resolved through discussion. Based on this literature search conducted by CADTH, no additional ITCs were included in this review report.

Description of Indirect Evidence

The NDS-MM-003 study (or KarMMa-RW), which is included in this indirect evidence section, is a comparative study of a retrospective body of RWE (a synthesized cohort) obtained from multiple databases of patients using available MM treatments to be compared to the single-arm KarMMa trial population treated with idecabtagene vicleucel. Details about the population, interventions, outcomes, and study characteristics are described in Table 18.

Methods

Objectives

The primary objective of the study was to describe demographic and selected clinical characteristics of RW patients with RRMM who received at least 3 prior myeloma regimens, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody (RRMM cohort), and to describe demographics, disease characteristics, treatment patterns, and clinical outcomes of the RW cohort and for the cohort of RW patients who met eligibility criteria for the KarMMa study (the eligible RRMM cohort).

As a secondary objective, investigators aimed to assess the comparative effectiveness of patients treated with idecabtagene vicleucel in the KarMMa study (MM-001) versus external RW controls (the balanced RRMM cohort) using a propensity score analyses.

Data Selection and Collection Methods

The organization, selection, and flow of patients from databases to build the different cohorts used in the study are shown in Figure 16 and described in the following section.

RRMM Cohort

Patients were first identified from RW databases (“real-world patients”) based on broad inclusion and exclusion criteria. For inclusion, patients had to be at least 18 years of age at the time of MM diagnosis; have a documented diagnosis of MM; have received a proteasome inhibitor, an IMiD, and an anti-CD38 antibody; have received anti-CD38 antibody therapy on or after November 16, 2015 (daratumumab earliest FDA approval); have received at least 3 prior MM treatment regimens by September 30, 2018; and have undergone at least 2 consecutive cycles of treatment for each regimen, unless PD was the best response to the regimen. Induction with or without hematopoietic stem cell transplant and with or without maintenance therapy was considered a single regimen.

Patients were excluded if they had previous exposure to any BCMA-directed therapy or gene-modified therapy.

Eligible RRMM Cohort

The eligible RRMM cohort was generated from the RRMM cohort described earlier using the following criteria:

• Additional inclusion criteria:

  • Refractory to the last regimen, with refractory defined as documented PD during or within 60 days (measured from the last dose) of completing treatment with the last regimen of AMT before study entry

  • Have measurable disease as measured by serum/urine M-protein and/or a serum FLC assay, including 1 of the following criteria:

    • serum M-protein greater or equal to 1.0 g/dL

    • urine M-protein greater or equal to 200 mg/24 hours

    • FLC level greater or equal to 10 mg/dL (100 mg/L) provided serum FLC ratio is abnormal

    • at least 1 documented disease assessment after the refractory condition or documented death.

• Exclusion criteria:

  • ECOG status of at least 2 at baseline (T₀, defined as the date that the RW subject became refractory to the last regimen)

  • Central nervous system involvement with MM

  • History or presence of clinically relevant pathology at the baseline time point (T₀), including CNS (e.g., epilepsy, seizure, paresis, aphasia, stroke, severe brain injuries, dementia, Parkinson disease, cerebellar disease, organic brain syndrome, or psychosis), unresolved hepatic pathology, unresolved cardiac, pulmonary, pathology, and inadequate renal or bone marrow function

  • Current or prior viral infection with hepatitis B virus, hepatitis C virus, or HIV

  • Solitary plasmacytomas without other evidence of measurable disease

  • Active or history of plasma cell leukemia

  • Secondary malignancies requiring therapy.

Matched RRMM Cohort

The eligible RRMM cohort served as a selection pool of RW patients to construct the matched RRMM cohort, a smaller cohort with baseline characteristics similar to those of the idecabtagene vicleucel cohort. Due to the small sample size of the eligible RRMM cohort, only a subset of the idecabtagene vicleucel cohort was matched to this cohort. The original statistical analysis plan specified an estimated 2:1 matched RRMM cohort of patients to compare real-world and idecabtagene vicleucel patients. However, due to strict criteria, fewer patients were available for the matched RRMM cohort, requiring a change in the statistical analysis. Investigators therefore decided to use trimmed stabilized IPTW, leaving the matched-paired and untrimmed IPTW as supportive analyses.

Idecabtagene Vicleucel Cohort (MM-001 – KarMMa Study)

The KarMMa study, the main intervention arm for the NDS-MM-003 study, is fully described in this CADTH clinical review report. Briefly, it is an ongoing, open-label, single-arm, multi-centre, phase II study to evaluate the efficacy and safety of idecabtagene vicleucel in patients with RRMM. The primary analysis population from the KarMMa study consisted of the 128 patients in the idecabtagene vicleucel–treated population, which is defined as all patients in the enrolled population who received idecabtagene vicleucel infusion (hereafter referred to as the idecabtagene vicleucel cohort); all efficacy results reported for the idecabtagene vicleucel cohort are therefore a composite from patients across the target dose levels of 150 × 10⁶ to 450 × 10⁶ CAR T cells used to treat patients.

Baseline, Index Date, and Study Period Definition

Baseline (T₀) for the eligible RRMM and matched RRMM cohorts was defined as the date that the RW subject became refractory to the last regimen, with refractory defined as documented PD during or within 60 days (measured from the last dose) of completing treatment with the last regimen of antimyeloma drug therapy before study entry.

Given that the RW subject met all eligibility criteria defined for each cohort, the index date or study day 1 was the start date of the first AMT after T₀. For patients in the RRMM cohort, day 1 was defined as the start date of a new treatment regimen after receiving at least 2 consecutive cycles of a proteasome inhibitor, an IMiD, and an anti-CD38 antibody therapy regimen by September 30, 2018, provided that anti-CD38 antibody therapy was received on or after November 16, 2015. For both the eligible RRMM cohort and matched RRMM cohort, day 1 was the start date of a new AMT after the occurrence of a PD that occurred within 60 days of completing the last qualified AMT, which can be the fourth or later regimen started after meeting study eligibility criteria.

For the idecabtagene vicleucel cohort, the baseline value was defined as the last value on or before the first dose date on which LDC was administered. The index date or study day 1 was the start date of idecabtagene vicleucel infusion after T₀.

Figure 16: Cohort Construction and Subject Selection in the KarMMa-RW Trial

ECOG = Eastern Cooperative Oncology Group; GRN = Guardian Research Network; MM = multiple myeloma; PD = progressive disease; PI = proteasome inhibitor; RRMM = relapsed and refractory multiple myeloma; T₀ = baseline time point.

^a Numbers (minimum to maximum) of matched patients from 30 imputed datasets.

Source: Sponsor-submitted indirect treatment comparison.¹²

Data Sources

Real-world data for NDS-MM-003 were acquired from 3 types of sources: clinical sites, the Connect MM Registry, and external research database partners. All RW patient-level data collection was retrospective and did not change clinical practice or patient visit schedules. Each data source was responsible for ensuring data collection followed applicable national and local ethical, legal, and privacy regulations.

Clinical sites data were obtained through manual chart abstraction into the study electronic case report forms. Individual clinical sites consisted of academic centres representing primarily North America and Europe. Out of 30 sites that were approached, 14 were initiated and able to complete data collection. All data collection was captured by an external company using a data collection instrument specifically designed for the purpose of this study, with audits and quality checks performed throughout the process.

The Connect MM Registry is a multi-centre, prospective observational cohort study designed to explore the natural history and real-world management of patients with newly diagnosed MM. Treatment selections and disease response assessments are performed at the discretion of the treating physician in accordance with their usual care. The registry follows patients from approximately 200 sites in the US with experience in oncology and/or hematology trials and registries and an adequate number of patients with MM.

Several external research databases are included. First of these is the COTA-RWE database of de-identified (US Health Insurance Portability and Accountability Act–compliant) patient-level electronic health record (EHR) data from contributing US academic, for-profit, and community oncologist provider sites and hospital systems. Also included are the Flatiron Health database, a database of EHR data from cancer clinics and unstructured reports from physicians; the Guardian Research Network, a nationwide consortium of health systems in the US that harness clinical and molecular oncology de-identified patient EHR information; |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

Data Collection, Integration, Quality Assurance, and Harmonization

Data from clinical sites were entered into a secure, internet-based electronic data-capture instrument by designated, trained personnel, with the principal investigator responsible for ensuring the accuracy and authenticity of the data.

Information from all sources was reformatted into a standardized study data model. Program and mapping specifications for variable transformations and derivations were created for each data element within each data source.

Upon receipt of raw data, a multi-stage quality control and validation protocol was initiated for each data source. The quality of raw data was evaluated through quantification of the magnitude and mechanism of missing data, density of laboratory results and clinical assessments, duplicated records, outliers, and extensive logic checks for time-related data points and those with dependencies. Upon completion of quality control and validation activities on the raw and integrated data, these data from multiple sources were pooled into a single dataset for harmonization.

End Points

The primary end point was the ORR, which was defined as the percentage of patients who achieved a PR or better. A PR was defined according to IMWG uniform response criteria for MM.³

Secondary end points included:

• The VGPR rate, which was defined as the percentage of patients who achieved VGPR or better from the index date according to IMWG uniform response criteria for MM.

• The CRR, which was defined as the percentage of patients who achieved CR or better according to IMWG uniform response criteria for MM. The assessment of this end point requires a bone marrow biopsy (also according to IMWG criteria), which is not typically performed in the routine care of RW patients, leading to limited data and the inability to assess CR for the RW population. The authors report this as “optional” and refer to analyses of VGPR or better to avoid underestimating response in the RW cohort.

• The OS, which was defined as the time from index date to time of death due to any cause. Patients who did not die were to be censored at the last date known to be alive or a maximum follow-up time of ||| days, whichever occurred first.

• The TTR, which was defined as the time from index date to first documentation of a response of PR or better in all patients responding.

• The DoR, which was defined for responders as time from first response (PR or better) to PD or death from any cause, whichever occurred first; responders who did not progress or did not die were to be censored on the last adequate assessment date.

• The PFS, which was defined as the time from index date to first documentation of PD or death due to any cause, whichever occurred first. Patients who did not have a PFS event were to be censored on the last adequate assessment date. Censoring rules based on EMA guidance for cancer trial end points were applied. Results using FDA censoring rules are also presented to assess any differences in both scenarios.^23,24

• Health care resource utilization, if available.

Analysis Methods

Descriptive analyses were performed to gain an understanding of the quality of the data collected and statistical distributions of characteristics of the RW patient cohorts. Continuous variables were described by the mean with standard deviation and 95% CI, median, upper, and lower quartiles, and maximum and minimum values. Categorical variables were reported as numbers and percentages with 95% CIs. Missing data were described in all summaries. For analysis comparing RW data with data from Study MM-001, a 2-sided P value of less than 0.05 was considered statistically significant without adjustment for multiplicity.

To enhance comparability between patients in the eligible RRMM cohort and in the idecabtagene vicleucel cohort (the KarMMa study), logistic regression was used to develop propensity scores, which were calculated on patient-level data to summarize the impact of covariates on treatment selection. These scores were then used by weighting individual patients (in the eligible RRMM cohort and the idecabtagene vicleucel cohort), IPTW, or through matching patients in these 2 cohorts to improve the balance of covariates and comparability between both groups. Authors used the propensity score trimmed stabilized IPTW method to enhance comparability of the baseline characteristics and weight the RW cohorts and the populations treated with idecabtagene vicleucel. These were achieved in 6 steps:

• Step 1. Selection of covariates to include in multivariate logistic regression modelling: Prior to starting the initial analysis, variables for consideration in the propensity score analysis were selected based on clinical importance relating to outcome and disease severity based on the expertise of the study team with a further review by a team of 7 clinical experts, who elected and ranked the order of prognostic factors across studies using a structured approach as recommended by guidance from the National Institute for Health and Care Excellence. A score was generated for each subject using a logistic model and adding further potential prognostic factors for efficacy to be considered in the propensity score model using the demographic and baseline characteristics from the KarMMa study and the eligible RRMM population. Univariate logistic regression was used to identify potential covariates for the initial propensity score model. Any values missing for the candidate covariates were imputed via multiple imputation procedures. Covariates considered included age, sex, bone lesions, time from initial diagnosis, number of prior regimens, cytogenetic high or low risk, refractory to immunomodulatory agents, refractory to proteasome inhibitors, refractory to anti-CD38 antibody, and baseline laboratory tests (platelet, hemoglobin, albumin, and calcium). Covariates that were considered important clinical outcome predictors by the scientific steering committee (i.e., age, albumin and number of prior regimens) were forced into the model.

• Step 2. Propensity score modelling: Once the covariates for the PS were calculated, 30 datasets were created (using multiple imputation of data when missing). Statistically significant variables (P values < 0.15) were included in the initial and/or full multivariate logistic regression to create a final propensity score model. This final model included age; baseline albumin; number of prior regimens; number of prior regimens/year; creatine clearance group; prior refractory to proteasome inhibitors, IMiDs, and anti-CD38 antibodies; bone lesions; and baseline calcium. A final propensity score for each imputed dataset and the stabilized IPTW were obtained and used to perform the balancing.

• Step 3. Assessment of balance between cohorts: Assessment of balance in PS between cohorts was conducted through a side-by-side comparison of raw baseline data for the eligible RRMM and the idecabtagene vicleucel cohort versus the balanced IPTW eligible RRMM and KarMMa baseline data. For these comparisons, pooled standardized mean differences were computed using Rubin’s rules before and after balancing and a threshold of 0.2 to indicate potentially important imbalances (Table 20).

• Step 4. Balancing methods and criteria: The balancing methodology used stabilized IPTW for the propensity score and greedy nearest-neighbour matching. The balancing originally anticipated that the eligible RRMM cohort would provide a 2:1 match for the idecabtagene vicleucel cohort, but both cohorts actually contained the same number of RW patients. The IPTW was chosen as the primary methodology, with a sensitivity analysis using the greedy nearest-neighbour matching (with a caliper width of 0.2 standard deviations of the logit of the propensity score). For 1:1 greedy nearest-neighbour matching, 1 subject in the RW cohort was matched with a single subject in the KarMMa study.

• Step 5. Estimation of risk ratios and hazard ratios for each of the 30 datasets mentioned in step 2.

• Step 6. Analysis with blinding (firewall) between the propensity score balancing process and the analysis of the post-baseline response and outcome data (blinded outcome assessors).

Because some imbalance remained between the 2 groups after propensity score balancing, doubly robust procedures, i.e., including the individual covariates from the initial propensity score model in addition to the IPTW in the analysis, were used to adjust for remaining unbalanced covariates for all the relevant comparisons. Due to the high degree of missing data (e.g., 46.8% of values in the RRMM eligible cohort were missing, as shown in Table 19), 30 separate datasets were generated using multiple imputation procedures and individual propensity scores were estimated within each dataset.

Several sensitivity analyses were conducted to assess the robustness of the primary analysis results. These included the use of the “greedy nearest-neighbour matching” (selected after a change in the statistical analysis plan, at which point the authors decided to use the IPTW as the primary methodology), analyses based on ECOG baseline data availability, and analyses using the enrolled population in the KarMMa study. Post hoc sensitivity analyses included those based on the European Union subgroup sensitivity analysis, which was used to contextualize the comparative evaluation of the primary outcome results within the European population, and a comparative analyses of the eligible RRMM cohort and the idecabtagene vicleucel–enrolled population for the ORR (primary end point) and VGPR.

The outcomes were reported as follows:

• The ORR was tabulated, together with the 2-sided 95% CI; estimates for the analyses were then obtained using Rubin’s rule to combine the individual estimates from each dataset. Relative risk, 95% CI, and the P value from the relative risk regression or equivalent generalized linear model were presented. A doubly robust procedure was used, including the covariates still differentiating these groups obtained before unblinding assessors to outcome data, in addition to the IPTW.

• The CR rate was tabulated with the 2-sided 95% CI and estimates for the analyses. The comparison of the eligible RRMM cohort to the idecabtagene vicleucel cohort was based on relative risk regression. Relative risk, 95% CI, and the P value from the generalized linear model were presented. A doubly robust procedure was used, including the covariates still differentiating these groups obtained before unblinding to outcome data.

• Overall survival (defined as time from index date to death due to any cause with patients censored on the last-known alive date or the maximum follow-up time of ||| days) was assessed using a Cox proportional hazards model for the time to event with the associated HR and 95% CI and P values. Adjusted survival curves were used to depict differences in survival over time between the eligible RRMM and the idecabtagene vicleucel cohorts with the same statistics reported as in the IPTW comparison, adjusted for differences in baseline covariates obtained in the final propensity score model (i.e., age; baseline albumin; number of prior regimens; number of prior regimens per year; creatine clearance group; prior refractory to propensity scores, IMiDs, and anti-CD38 antibodies; bone lesions; and baseline calcium).

• The TTR was summarized using descriptive statistics for the eligible RRMM cohort and the idecabtagene vicleucel cohort.

• The DoR comparison was conducted using KM graphics, a Cox proportional hazards model, and the associated estimable group medians, HR with 95% CI and P value for differences. Survival plots and the associated statistics described earlier, adjusted for covariates (age; baseline albumin; number of prior regimens; number of prior regimens/year; creatine clearance group; prior refractory to proteasome inhibitors, IMiDs, and anti-CD38 antibodies; bone lesions; and baseline calcium), were used to compare DoR in the eligible RRMM cohort and the idecabtagene vicleucel cohort.

• Progression-free survival was assessed in patients who had not progressed and were still alive according to EMA censoring rules (and presenting based on FDA rules) on the last adequate assessment date or at 730 days, whichever occurred first. A PFS comparison of the cohorts was conducted using KM graphics, Cox proportional hazards model, and the associated estimable group medians, HR with 95% CI, and P value for differences. Adjusted survival curves were used to depict differences in PFS over time between the eligible RRMM and the idecabtagene vicleucel cohorts with the same statistics reported as in the matched comparison, adjusted for differences in baseline covariates (i.e., age; baseline albumin; number of prior regimens; number of prior regimens/year; creatine clearance group; prior refractory to proteasome inhibitors, IMiDs, and anti-CD38 antibodies; bone lesions; and baseline calcium).

Results

Disposition

In total, 1,949 RW patients were included in the broad RRMM cohort, and of these, 1,171 were refractory to the last treatment regimen at baseline. Of the 1,171 patients in the refractory RRMM cohort, 643 did not receive a next-line treatment (441 due to death and 202 due to lack of follow-up data). This resulted in 528 patients in the RRMM cohort who received new therapy. Additional eligibility criteria similar to those of Study MM-001 were then applied, resulting in the selection of 190 RW patients for the eligible RRMM cohort.

For the KarMMa study, 140 patients underwent leukapheresis and were included in the enrolled population. Of these, 128 patients received idecabtagene vicleucel treatment and were included in the idecabtagene vicleucel cohort as the primary analysis population for comparison against the eligible RRMM cohort.

Baseline demographics and disease characteristics of the RRMM cohort, the eligible RRMM cohort, and the idecabtagene vicleucel population are presented in Table 19, while Table 20 lists the variables before and after balancing (using the trimmed stabilized IPTW method) of the eligible RRMM and idecabtagene vicleucel cohorts are shown.

A total of 108 (56.8%) patients in the eligible RRMM cohort had discontinued from the study, all due to death. As of the October 30, 2019, data cut-off date, 58 patients (45.3%) in the idecabtagene vicleucel cohort had discontinued, with 31 patients (24.2%) discontinuing due to death, 26 (20.3%) discontinuing due to withdrawal by subject, and 1 (0.8) patient discontinuing due to lost to follow-up.

In the eligible RRMM cohort, the median duration of follow-up for all treated patients (which includes duration up to death for patients who died and duration up to last date known alive for surviving patients), was 10.2 months (range = 0.2 to 24.0). In the idecabtagene vicleucel cohort, the median duration of follow-up after idecabtagene vicleucel infusion was 11.3 months (range = 0.2 to 18.6). These numbers were current as of the October 30, 2019, data cut-off date of the KarMMa study.

The median follow-up duration for surviving patients (n = 82) was 15.0 months (range = 0.8 to 24.0) in the eligible RRMM cohort, while in the idecabtagene vicleucel–treated population (n = 94) the median was |||| months (range = |||||||||||||).

The most common regimens (i.e., those received by ≥ 5 patients) in the eligible RRMM cohort are listed in Table 21.

Results

Overall Response Rate

As of the data cut-off date, the ORR was lower in the eligible RRMM cohort compared with the idecabtagene vicleucel cohort (32.2% versus 76.4%; risk ratio = 2.4; 95% CI, 1.7 to 3.3; P < 0.0001). A summary of the ORR and VGPR-or-better rate adjusted for trimmed stabilized IPTW for patients in the eligible RRMM and idecabtagene vicleucel cohorts is presented in Table 22. Results of the ORR and VGPR-or-better rate from the supporting sensitivity analyses in the idecabtagene vicleucel–enrolled population were similar to those observed for the idecabtagene vicleucel–treated cohort used for the primary analysis.

The assessment of a CR according to IMWG criteria requires a bone marrow biopsy to be performed for patients. As a bone marrow biopsy is not typically performed in the routine care of RW patients, the CR rate was unable to be analyzed due to the limited availability of bone marrow biopsy results for the eligible RRMM cohort.

Overall Survival

After a median follow-up time of 15 months for all surviving patients in the eligible RRMM cohort, 55.8% of patients had died, while after a median follow-up time of ||| months for all surviving patients in the idecabtagene vicleucel cohort, a total of |||||| of patients had died (Figure 17). The median OS for RW patients in the eligible RRMM cohort was 14.7 months (95% CI, ||||||||||||||) and was |||||||||||| (95% CI = NE) in the idecabtagene vicleucel cohort. A comparison between the 2 groups yielded an HR for OS of ||||||||||||||||||||||||||||||||, favouring the idecabtagene vicleucel cohort compared with the eligible RRMM cohort treated with available therapy (P = |||||||).

Figure 17: [Redacted]

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

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

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

The figure was redacted as per the sponsor’s request.

Time to Response

As of the data cut-off date, the median TTR for responders was 1.1 months (range = 0.2 to 8.6) in the eligible RRMM cohort versus 1.0 month (range = 0.5 to 8.8) in the idecabtagene vicleucel cohort. As of the data cut-off date, the median DoR was 9.0 months (95% CI, 7.5 to ||||) in the eligible RRMM cohort versus ||| months (95% CI, ||||||||||||||) in the idecabtagene vicleucel cohort (Table 23).

Progression-Free Survival

This evaluation was made based on EMA censoring rules; however, by using an FDA censoring scenario, the results were comparable. As of the data cut-off date, the median PFS adjusted by the trimmed stabilized IPTW for comparability was lower in the eligible RRMM cohort compared with the idecabtagene vicleucel cohort (3.5 months versus 11.1 months, respectively; Table 24).

After a median follow-up time of || | months for all RW patients in the eligible RRMM cohort, || || of patients had a PFS event (PD or death), while after a median follow-up time of || | months for all patients in the idecabtagene vicleucel cohort, | ||| of patients had a PFS event. A comparison between the 2 groups yielded an HR for PFS of |||||||||||||||||||||||||||||||, favouring the idecabtagene vicleucel cohort compared with the eligible RRMM cohort treated with available therapy (||||||||||||; Figure 18).

Figure 18: [Redacted]

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

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

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

The figure was redacted as per the sponsor’s request.

An evaluation of the ORR outcomes by subgroup revealed that, for all subgroups evaluated, the results were consistent with the overall results for the ORR. In all subgroups in which the ORR was evaluated the benefit in the idecabtagene vicleucel cohort was maintained compared with the eligible RRMM cohort. The results were similar for the PFS outcome.

Critical Appraisal of NDS-MM-003

The NDS-MM-003 study compared a single-arm clinical trial (KarMMa) against a matched RRMM cohort from different databases and clinical sites. Investigators planned to achieve a balance of prognostic factors between these 2 groups for comparison purpose by using propensity score models. Due to the small sample size obtained from the databases (with important missingness of data), the investigators changed the methodology to restrict the primary analysis to an eligible RRMM cohort, and balanced factors between the eligible RRMM cohort and the idecabtagene vicleucel–treated population from the KarMMa study using the trimmed stabilized IPTW method, leaving the matched cohort evaluation as supportive analyses.

From the databases, the baseline disease characteristics were generally similar among RW patients from the different data sources. Patients from IQVIA, the Guardian Research Network, and M2GEN tended to be less heavily pretreated, with a lower number of prior antimyeloma regimens. Patients from IQVIA also had a lower degree of refractoriness to prior AMTs, as evidenced by a lower percentage of patients who were double-class or triple-class refractory.

The missing data from these databases throughout the process of creating the eligible cohort implies a potential bias due to the need to rely on multiple imputation methods. More than half of the data for a number of variables missing would have been important to include in the propensity scores. Multiple imputations of missing values made in generating the eligible RRMM cohort can increase the uncertainty in effect estimates. Furthermore, no methods were used to control the type I error rate across multiple outcomes, subgroups, or analyses.

Although later addressed in the propensity score balancing, baseline differences in multiple variables observed between the RRMM cohort, the eligible RRMM cohort, and the idecabtagene vicleucel cohort (e.g., the presence of plasmacytoma or refractory status) suggest the possibility of other unmeasured confounding factors, and the prospect that these can be unevenly distributed between groups without proper randomization. Prior to the IPTW adjustment, with the exception of age, the imbalance of the prognostic factors and prior treatments suggests the eligible RRMM cohort was a less heavily pretreated and less refractory population with a more favourable prognosis overall compared with the idecabtagene vicleucel cohort. Furthermore, differences in study designs and combining patients across databases (i.e., selecting patients from different databases) may introduce heterogeneity that cannot be accounted for in propensity score modelling. There is also the potential for differences in available covariates across databases, definitions of covariates, and outcome assessments due to practice variation across the regions and centres where the databases capture patients. Similarly, there is the potential for heterogenous assessment of end points when comparing an assessment by an IRC versus a clinician assessment as it was made in the KarMMa study.

Overall, 190 of the 1,171 patients of the RRMM cohort were included in the final set of eligible RRMM patients compared with the idecabtagene vicleucel cohort, which can be considered a significant reduction.

Further critical appraisal points include a lack of clarity regarding how the retrospective assessment of the IMWG for response was determined for the RWE cohorts (as opposed to the KarMMa study). Blinding of analysts to generate propensity scores occurred only for the first wave of analysis in this study, and a second wave of analysis was performed in an unblinded fashion, which can generate bias in outcome assessors. Clinical relevance should be considered to select the covariates used in propensity score models; investigators used, in part, a data-driven approach, which is not ideal. No data were available for health care resource utilization and drug exposure. It was not clear at what time points the interventions were available in the different databases; this can add generalizability issues, as rapidly changing practices mean some regimens may be outdated, introducing heterogeneity to the interventions.

Although the population from the eligible RRMM cohort is similar to the idecabtagene vicleucel cohort in several measures, generalizability of this and the matched final cohort as well as the comparisons can be a concern in the Canadian clinical practice due to the different and varied therapies included in the regimens provided. The availability of such regimens and drugs will vary depending on funding and feasibility for each province/territory. Only a fraction of the regimens evaluated as comparators in this indirect comparison were described by the clinical experts consulted by CADTH as commonly used in clinical practice in Canada. These regimens include pomalidomide plus dexamethasone, daratumumab plus dexamethasone plus lenalidomide, and Cd. The individual medications are approved in Canada. With regards to external validity, the RWE was derived mainly from retrospective databases within the US. It is likely a number of important differences between the US and Canada in the management of these patients would be expected to affect outcomes.

Summary

The NDS-MM-003 presented in this ITC section is a comparison of the patients from the single-arm, open-label KarMMa study who were treated with idecabtagene vicleucel (in a dose range of 150 × 10⁶ to 450 × 10⁶ CAR T cells) with a cohort of real-world RRMM patients obtained from different patient-level clinical sites or databases that function as the real-world comparison arm. Patients obtained from these cohorts of clinical sites and registries were selected to be similar to those in the KarMMa study, i.e., triple-class exposed to IMiD, proteasome inhibitors, and anti-CD38 antibodies. Authors of the study identified prognostic variables and adjusted for some of these using an IPTW methodology to make the groups more comparable. Due to a smaller-than-anticipated sample size, authors focused their comparison of the KarMMa trial and the eligible RRMM cohort using the trimmed stabilized IPTW method and presented the matched RRMM cohort comparison as supportive analysis.

The primary analysis focused on the ORR after adjustment of eligibility criteria, in which patients in the idecabtagene vicleucel cohort showed an improved ORR. Similarly, patients in the idecabtagene vicleucel arm had better PFS and OS after a median of 15 months of follow-up. No data on AEs, HRQoL, or health care resource utilization were provided.

While this propensity score analysis provides indirect evidence of the comparative effectiveness of idecabtagene vicleucel in patients who are refractory to multiple AMT, the evidence is limited by inherent design differences between the studies that cannot be adjusted for statistically, the potential impact of unmeasured and unaccounted-for prognostic factors and effect modifiers in the models, and reduced generalizability due to the inclusion of irrelevant comparators.

Other Relevant Indirect Treatment Comparisons

This section includes a sponsor-submitted ITC assessed as a high-level summary of information of the MAIC of efficacy outcomes based on the BB2121-MM-001 (KarMMa) and MAMMOTH studies.

Description of the Study

Based on the information from the KarMMa trial, which enrolled patients who have received at least 3 prior AMT regimens, been refractory to the last regimen received, and been previously treated with an IMiD, a proteasome inhibitor, and an anti-CD38 antibody, the authors of the study compare the effect of idecabtagene vicleucel single-arm trial data against independent data from observational studies (not funded by the sponsor) to provide further RWE for decision-makers and stakeholders. The largest such real-world study, MAMMOTH,³⁴ was chosen by the authors as it evaluates conventional care in 275 patients with MM in US medical centres who were refractory to daratumumab or isatuximab, administered alone or in combination.

The current indirect comparison aimed to evaluate the comparative efficacy of idecabtagene vicleucel versus conventional care in terms of the ORR, as well as PFS and OS, for the treatment of patients with RRMM who have received 3 or more prior regimens, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody, using an unanchored MAIC of KarMMa and MAMMOTH.

The unanchored MAIC uses KarMMa data with a cut-off date of |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. The median duration of follow-up after idecabtagene vicleucel infusion for all idecabtagene vicleucel–infused patients was |||||||||||||. This evidence on the clinical efficacy of idecabtagene vicleucel was obtained directly from the Clinical Study Report.

The authors of the unanchored MAIC performed a literature search that found 11 real-world studies, from which they selected MAMMOTH as the study with the largest sample size and available KM data.

MAMMOTH is a large, retrospective chart review of observational data of 275 RRMM patients from the US who are refractory to daratumumab or isatuximab, administered alone or in combination. Of these patients, 249 received a variety of regimens corresponding to conventional care following progression on anti-CD38 antibodies (treated population). The 249 patients in the treated population were categorized based on the individual conventional care regimens they received.

The following treatment groups correspond to the most common interventions received by patients in the NDS-MM-003 (RWE) study¹² previously mentioned in this CADTH report, and which were selected by the investigators as the intervention subgroups:

• daratumumab plus an IMiD (41 patients)

• elotuzumab plus an IMiD (19 patients)

• carfilzomib plus an alkylator (19 patients)

• PACE-like (24 patients)

• daratumumab plus a proteasome inhibitor (13 patients)

• carfilzomib plus an IMiD (34 patients).

Data for the MAMMOTH study was extracted from its primary publication.³⁴ MAMMOTH included patients who had received an anti-CD38 antibody therapy as their last line of treatment and were refractory to it. KarMMa restricted patients to those with an ECOG score of 0 or 1, while MAMMOTH did not have any ECOG restrictions.

The primary outcome, ORR, was defined in the KarMMa study using the IMWG criteria defined by the IRC, as well as in MAMMOTH, although in the latter study the ORR was calculated by combining the PR and better-than-VGPR rates. Definitions for PFS events were similar, and disease progression was defined based on IMWG response criteria in both studies. In the MAMMOTH study, PFS was measured as the onset of the next line of treatment in the treated population (PFS was not reported for the overall population). While OS was measured from either the time of leukapheresis (enrolled population) or the time of idecabtagene vicleucel infusion (infused population) in KarMMa, it was calculated from the time of PD on daratumumab or isatuximab for patients in MAMMOTH for all patient groups, and therefore includes a period of time during which the patient is untreated before they begin their next line of treatment.

Methods

This is an unanchored MAIC between idecabtagene vicleucel and mixed comparators using RWE from the MAMMOTH study. Because individual patient data for the comparator study (MAMMOTH) was not available, the analyses were limited to the study-level aggregate data from the publication.

Analyses were performed for the KarMMa populations (enrolled and infused populations as well as a target dose of 450 × 10⁶ CAR T cells) versus the MAMMOTH treated population in terms of ORR, PFS, and OS. In MAMMOTH, the 249 patients in the treated population were categorized to 6 subgroups related to the interventions described earlier.

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

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

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

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

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

Efficacy Results

Patients in KarMMa tended to be younger than those in MAMMOTH, whereas the distribution of gender and race was similar in the 2 studies. Patients in KarMMa had the disease for a slightly longer time compared to MAMMOTH. In terms of prior treatment experience, patients in KarMMA and MAMMOTH had a median of 6 prior regimens versus 5 prior regimens, respectively. The proportions of patients refractory to bortezomib and those refractory to lenalidomide were also similar between the 2 studies, whereas a slightly higher percentage of patients were refractory to carfilzomib in KarMMa, as was the case for refractoriness to pomalidomide. Last, while the vast majority of patients had previously received autologous stem cell transplantation in KarMMa, the proportion of such patients was lower in MAMMOTH. Overall, patients were more refractory to prior treatments in KarMMa.

The KarMMa study included more heavily pretreated and refractory patients with a less-favourable prognosis compared with the MAMMOTH patients.

The authors mention effective sample size (EES) estimates in the enrolled and infused populations that were similar to existing ESS estimates in the literature, both in absolute terms and with respect to relative reductions (||||||||| for both the enrolled and infused populations when matched to the treated population in MAMMOTH). From literature evaluated by the authors, ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

The baseline characteristics are presented in Table 25.

Overall Response Rate

Idecabtagene vicleucel in the enrolled population was more efficacious than conventional care in terms of ORR, whether the analysis was naive or adjusted using the MAIC (||||||||||||||||||||||||||||||||||||||||||||||||). These results were consistent with the KarMMa infused population (||||||||||||||||||||||||||||||||||||||||||||||||||||||).

Figure 19: [Redacted]

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

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

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

The figure was redacted as per the sponsor’s request.

| ||

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

| |

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

Figure 20: [Redacted]

|||||||||||||||||||||||||||||||||¹³

Note: This figure has been redacted as per the sponsor’s request.

Figure 21: [Redacted]

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

Note: This figure has been redacted as per the sponsor’s request.

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

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

Harms Results

No data about harms were reported.

Critical Appraisal

This is an unanchored MAIC with an appropriate rationale for comparing individual patient data from a single-arm study with aggregate data from a published study to assess the effect of idecabtagene vicleucel treatment in a population with RRMM.

The approach to identification of prognostic factors and effect modifiers to include in the MAIC (i.e., a literature review designed to identify prognostic factors, a series of interviews with clinical experts, and an analysis described in this report from CADTH comparing MM-001 to MM-003) was appropriate. However, it is not clear if the underlying assumption of the unanchored MAIC that all effect modifiers and prognostic factors have been accounted for was accomplished. It is unclear whether all of the relevant differences in patient characteristics were captured (for both the effect modifiers and prognostic factors), as a number of identified prognostic elements were not reported in MAMMOTH. The ESS estimates for reduction were considered close to what is published in the literature related to oncology trials. For time-to-event outcomes (PFS and OS) the proportional hazards assumption was evaluated with resulting potential violations, and any HR should be interpreted with caution.

The MAIC cannot address heterogeneity in study designs. As a comparison between clinical trial data and RWE, differences in outcome assessments and definitions (e.g., differences in setting and follow-ups) can generate uncertainty. The analysis is limited to study-level aggregated data from the full-text publication for MAMMOTH and compared to the KarMMa study.

The different treatments obtained in the MAMMOTH study also generate uncertainty about the generalizability of the results, particularly from the Canadian perspective, due to the fact that approximately a third of the comparators are not widely available in the different provinces and territories.

Overall, beneficial effect estimates were observed in favour of idecabtagene vicleucel, although the certainty in these estimates remains unclear due to the observational nature of the evidence base, the possibility of residual confounding factors, heterogeneity, and risk of bias.

Discussion

Summary of Available Evidence

One clinical study was included in this report. The KarMMa study (N = 140) was a phase II, open-label, single-arm, prospective trial. Eligible patients were those who have received multiple previous treatments for RRMM, including an IMiD, a proteasome inhibitor, and an anti-CD38 antibody. A total of 128 patients received idecabtagene vicleucel target doses of 150 × 10⁶, 300 × 10⁶, or 450 × 10⁶ CAR T cells. The primary end point was the ORR (a PR or better). The key secondary end point was a CR or better (comprising CR and sCR). Other secondary end points included DoR, TTR, TTP, MRD-negative status, survival outcomes (OS and PFS), and HRQoL.

In addition, 2 sponsor-submitted ITCs were included to provide indirect evidence of the comparative effectiveness of idecabtagene vicleucel to relevant treatment alternatives. The first is a propensity score analysis comparing the KarMMa trial MM-001 single-arm study to pooled RWE obtained from multiple clinical sites and registries from Europe and the US. The second ITC is a MAIC that compares the individual patient data from the KarMMa trial to aggregate-level data from a published RWE study that included multiple comparators. Both approaches aim at providing a comparator arm, with adjustment for differences in baseline characteristics, prognostic factors, and effect modifiers to produce an adjusted effect estimate for decision-makers and stakeholders, given the lack of a direct, head-to-head comparison of idecabtagene vicleucel to relevant comparators.

Interpretation of Results

Efficacy

The ORR, CRR, DoR, OS, and HRQoL were considered the most important outcomes from the perspective of clinical experts and clinician groups. In addition, patient groups considered improved HRQoL domains (fewer symptoms, higher functionality, and fewer side effects from treatments) to be highly valued outcomes.

At a median follow-up of 11.3 months, the KarMMa trial met its primary end point for the target dose of 450 × 106 CAR T cells, achieving an ORR of 81.5%, which was above the null hypothesis of 50% established in the protocol. Patients reached an ORR above this point in all target dose groups. Given that the test for the ORR (primary end point) was positive, the key secondary efficacy end point of the CRR was tested against the prespecified null hypothesis of 10% or lower. The result rejected the null hypothesis, with 35.2% of patients achieving a CR or better. Both these values from the primary and key secondary end points were considered clinically meaningful by the clinical experts.

The DoR, TTR, and TTP were evaluated in the KarMMa trial as secondary end points. With a median 10.6 months for DoR, 1 month for TTR, and 9 months for TTP, the results suggest a beneficial effect in patients treated with idecabtagene vicleucel and were considered by the clinical experts likely to represent a meaningful impact. Likewise, the MRD-negative status with a CR or better observed in a quarter of the patients treated with idecabtagene vicleucel was also supportive of a clinical benefit. However, none of these secondary end points were adjusted for multiplicity and all should be interpreted with caution.

Overall survival was considered by clinical experts and patients to be among the most valued outcomes used in clinical trials including those involving patients with MM. The observed median OS of 18.2 months, with 76% of individuals alive at 12 months, was considered by the clinical experts to be an overall beneficial effect estimate for patients with the characteristics included in the KarMMa trial (i.e., those with a median of 6 years since diagnosis, with relapses, and who have been heavily treated). Similarly, based on the KM estimates, the median PFS of 8.6 months, with 34% of patients event-free at 12 months, denoted a likely meaningful benefit for patients treated with idecabtagene vicleucel.

Health-related QoL outcomes were frequently mentioned in input from patient and clinician groups. Patients put a higher value on decreasing symptoms, improved physical functioning, less pain, and fewer side effects. In the KarMMa trial, idecabtagene vicleucel treatment was associated with improvements in the fatigue, pain, physical functioning, and global health/QoL subscales of the EORTC QLQ-C30 by reaching points of meaningful significance above the thresholds of probable benefit, according to MIDs established in the literature. On average, no clinically meaningful deterioration in the EORTC QLQ-C30 cognitive functioning and EORTC QLQ-MY20 disease symptoms and side effects subscales were observed posttreatment. With respect to the EQ-5D-5L subscales, no changes from baseline were observed in all subscales of this measurement. These findings suggest that idecabtagene vicleucel treatment is associated with improvements or stabilization in most primary HRQoL metrics. However, these results were observed over the first and middle parts of the study follow-up, eventually decreasing to baseline levels, with added uncertainty due to imprecision (i.e., fewer patients at the end of the study) and the fact that all HRQoL outcomes in the KarMMa study were secondary end points assessed without adjustment for multiplicity.

Evaluation of different scenarios (sensitivity analyses) in the analysis of the outcomes revealed no notable differences that had an impact on the final effect estimates. The investigator therefore obtained estimates based on whether the FDA or EMA rules for censoring were used (DoR, TTP, and PFS outcomes), or whether comparisons used IRC or investigator adjudication of end points (for ORR, CRR, and TTP). Finally, scenarios were presented comparing the enrolled population against the population treated with idecabtagene vicleucel (for DoR, CRR, PFS, and OS).

With the exception of the ORR and CR outcomes, no other secondary end points were considered in the adjustment for multiplicity, which should be interpreted with consideration of type I error.

Harms

All 128 patients treated with idecabtagene vicleucel (on or after the idecabtagene vicleucel infusion) reported AEs. The most common were neutropenia (91%), anemia (70%), and thrombocytopenia (63%); these were described in the patient group input as important but of low value when weighed against likely benefits, such as longer survival, remission, HRQoL, and reduced burden of symptoms. The clinical experts consulted by CADTH indicated that these AEs were expected, and the patient group provided similar input.

Serious AEs on or after idecabtagene vicleucel infusion were common (reported by almost two-thirds of patients) and comprised infections (pneumonia, sepsis), CRS, and blood disorders (febrile neutropenia, neutropenia, and thrombocytopenia).

Deaths on or after idecabtagene vicleucel infusion occurred in 34 patients as of the data cut-off date. Most deaths were attributed to the disease under study or complications attributed to the disease under study.

Among the notable harms established in this review protocol, CRS was common, as it was reported in 107 patients (84%). This was in accordance with the expectations from the clinical experts consulted by CADTH, who highlighted the need for treatment for this condition. Although CRS is manageable, it can impose extra costs and burdens on patients that must be weighed against the desirable effects of the therapy.

Neurotoxicity in the KarMMa trial was defined as investigator-identified neurotoxicity. These developed in 23 patients (18%), and included several signs and symptoms, such as a confusional state and encephalopathy. These symptoms and their frequency were important to both experts and patients.

No concerns were expressed over hospitalization rates or hospital utilization based on the number of patients admitted to ICUs during the study. Overall, the impact of the numbers of ICU admissions in the KarMMa trial was not an issue for patient groups or clinical experts.

No data on harms were provided from the RWE propensity score comparisons or ITC (MAIC) evaluated in this review.

Conclusions

Evidence from the single-arm, open-label KarMMa-001 study suggests that the CAR T-cell therapy idecabtagene vicleucel at a target dose of 450 × 10⁶ CAR T cells is associated with an induction of response, based on an ORR of 81.5% and a CR of 35.2%. The results appear to indicate that treatment with idecabtagene vicleucel may offer a benefit in terms of DoR, MRD status, TTR, TTP, OS, and PFS, but none of these outcomes were controlled for multiplicity so the results should be interpreted with consideration of type I error. The data for HRQoL outcomes were descriptive and could not be interpreted due to missing data. These effect estimates come from evidence with risk of bias due to the lack of estimations from a comparison arm in a randomized design, and must be weighed against the observed undesirable effects, including AEs (such as CRS), the need for LDC, and bridging therapy. All patients treated with idecabtagene vicleucel reported at least 1 AE. The most commonly reported AEs were neutropenia, CRS, anemia, and thrombocytopenia.

Results of 2 indirect comparisons of the population from the KarMMa study to individual patient observational data (RWE) and aggregated data from published studies add support for the beneficial effects observed in KarMMa, but also include a risk of bias due to inherent design differences in the bodies of evidence that cannot be adjusted for statistically, the potential impact of unmeasured and unaccounted-for prognostic factors and effect modifiers in the models, and undermined generalizability by the inclusion of irrelevant comparators. Given the limitations of the 2 ITCs and the absence of direct comparative evidence, any potential benefit of idecabtagene vicleucel compared with other treatment regimens used in this patient population remains unknown.

References

1.Tsang M, Le M, Ghazawi FM, et al. Multiple myeloma epidemiology and patient geographic distribution in Canada: a population study. Cancer. 2019;125(14):2435-2444. PubMed

2.Rajkumar SV, Kumar S. Multiple myeloma: diagnosis and treatment. Mayo Clin Proc. 2016;91(1):101-119. PubMed

3.Kumar S, Paiva B, Anderson KC, et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 2016;17(8):e328-e346. PubMed

4.Moreau P, San Miguel J, Ludwig H, et al. Multiple myeloma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24 Suppl 6:vi133-137. PubMed

5.Brenner DR, Weir HK, Demers AA, et al. Projected estimates of cancer in Canada in 2020. CMAJ. 2020;192(9):E199-E205. PubMed

6.Dimopoulos MA, Weisel K, Kaufman J, et al. A systematic literature review and network meta-analysis evaluating the efficacy of daratumumab-based regimens in patients with relapsed/refractory multiple myeloma. Indian J Hematol Blood Transfus. 2017;33 (1 Supplement 1):S2.

7.Mikhael J. Treatment options for triple-class refractory multiple myeloma. Clin Lymphoma Myeloma Leuk. 2020;20(1):1-7. PubMed

8.Mikhael J, Ismaila N, Cheung MC, et al. Treatment of multiple myeloma: ASCO and CCO joint clinical practice guideline. J Clin Oncol. 2019;37(14):1228-1263. PubMed

9.Moreau P, Kumar SK, San Miguel J, et al. Treatment of relapsed and refractory multiple myeloma: recommendations from the International Myeloma Working Group. Lancet Oncol. 2021;22(3):e105-e118. PubMed

10.Drug Reimbursement Review sponsor submission: Abecma (idecabtagene vicleucel), single dose cell suspension for intravenous administration[internal sponsor's package]. St-Laurent (QC): Celgene Inc., a Bristol Myers Squibb company; 2020 Dec 16.

11.Clinical Study Report: Protocol BB2121-MM-001. A phase 2, multicenter study to determine the efficacy and safety of bb2121 in subjects with relapsed and refractory multiple myeloma [internal sponsor's report]. Summit (NJ): Celgene Corporation; 2017 Aug 25.

12.NDS-MM-003: A global, non-interventional, retrospective, multi-center study to generate real-world evidence of subjects with relapsed and refractory multiple myeloma with prior exposure to an anti-cd38 antibody [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Abecma (idecabtagene vicleucel), single dose cell suspension for intravenous administration. Summit (NJ): Celgene Corporation; 2020 Mar 16.

13.A matching-adjusted indirect treatment comparison of efficacy outcomes based on BB2121-MM-001 (15+1) and MAMMOTH pivotal studies [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Abecma (idecabtagene vicleucel), single dose cell suspension for intravenous administration. Saint Laurent (QC): Celgene Inc; 2020.

14.Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136(5):E359-386. PubMed

15.Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15(12):e538-548. PubMed

16.Nijhof IS, van de Donk N, Zweegman S, Lokhorst HM. Current and new therapeutic strategies for relapsed and refractory multiple myeloma: an update. Drugs. 2018;78(1):19-37. PubMed