Reimbursement Review

Ciltacabtagene Autoleucel (Carvykti)

Sponsor: Janssen Inc.

Therapeutic area: Relapsed or refractory multiple myeloma

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Ethics Review

Clinical Review

Abbreviations

Executive Summary

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

Table 1: Background Information of Application Submitted for Review

CAR = chimeric antigen receptor; NOC = Notice of Compliance.

Introduction

Multiple myeloma (MM) is a hematological malignancy characterized by clonal proliferation of malignant plasma cells (B-cells) driven by an oncologic event and consequent overproduction of the abnormal immunoglobulin monoclonal protein (M protein).¹ The estimated number of newly diagnosed cases of MM in Canada was 4,000 in 2022 and 3,900 in 2023.^2,3 Based on the reported MM prevalence in 2018 and the growing projected annual incidence rate, combined with a predicted 5-year survival rate, the projected prevalence of MM is estimated to be approximately 17,568 in Canada (excluding Quebec) in 2025.⁴ The majority of patients with MM will relapse, and many patients will develop disease that is refractory to commonly used therapies. Patients with relapsed or refractory MM (RRMM) often undergo multiple rounds of treatment, with the duration of remission, depth of response, progression-free survival (PFS), and overall survival (OS) decreasing with each subsequent line of therapy.^5-8 According to the clinical experts and clinician groups, the main treatment goals for patients with RRMM are to prolong survival, improve symptoms, minimize toxicities, and improve health-related quality of life (HRQoL).Therapies for the treatment of patients with RRMM, and the sequencing of these treatments, depends on eligibility for autologous stem cell transplant at diagnosis, patient age, comorbidities, previous treatments, beforexicities, and line of therapy. Available treatment options for patients with RRMM in Canada include triplet therapy — consisting of proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), or monoclonal antibodies (mABs) — and chimeric antigen receptor (CAR) T-cell therapy (i.e., ciltacabtagene autoleucel [cilta-cel], which is under consideration for negotiation at the pan-Canadian Pharmaceutical Alliance).^9,10

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of cilta-cel, cell suspension in infusion bag, 0.5 × 10⁶ to 1.0 × 10⁶ CAR-positive viable T cells per kilogram of body weight, with a maximum of 1 × 10⁸ CAR-positive viable T cells, for IV infusion in the treatment of RRMM in adult patients.

Cilta-cel was previously reviewed by CADTH for the treatment of adult patients with RRMM who have received at least 3 prior lines of therapy, including a PI, an IMiD, and an anti-CD38 mAB, and whose disease is refractory to their last treatment.¹¹

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of the input provided by the patient and clinician groups who responded to the Canada’s Drug Agency (CDA-AMC) call for input and from clinical expert(s) consulted for the purpose of this review.

Patient Input

CDA-AMC received 1 patient group submission, from Myeloma Canada. Myeloma Canada is a patient advocacy group that supports patients diagnosed with MM living in Canada.

Myeloma Canada collected data from patients with RRMM who had received 1 to 3 prior lines of therapy and whose disease was refractory to lenalidomide or who had experience with a CAR T-cell therapy, or from the caregivers of such patients, through a survey across Canada and internationally, via email and social media from April 5, 2024, to April 24, 2024. There were 53 eligible respondents; 51 lived in Canada (Alberta [3], British Columbia [12], Newfoundland and Labrador [2], Ontario [29], Quebec [5]), and 2 lived in France. There were 2 subsets of survey respondents: 1 subset comprised 37 patients or caregivers who met the criteria for the indication under review, and the other comprised 16 respondents who had CAR T-cell therapy experience, of which 8 patients or caregivers had experience with cilta-cel and 8 patients or caregivers had experience with a different CAR T-cell therapy.

In terms of MM disease complications, infections were considered the most important aspect to control, followed by kidney problems. Patients and/or caregivers also reported that MM had various impacts on their quality of life, such as limiting their ability to travel and their pursuit of life goals or accomplishments. Most patients and caregivers identified a need for effective MM treatment options, with manageable side effects and minimal impact on quality of life. Of the 37 patients or caregivers who met the criteria of the indication under review, 22 reported receiving 3 lines of therapy and 2 reported treatment with B-cell maturation antigen–targeted therapy. The experiences shared by patients or caregivers who received CAR T-cell therapy were generally positive. Of the 8 respondents who received cilta-cel, 5 rated the treatment extremely effective and the side effects extremely tolerable. Cytokine release syndrome (CRS) was perceived to be the most concerning side effect by patients who met the criteria for the indication under review. However, it was considered bearable for respondents who had received cilta-cel. Twenty-eight respondents out of the 37 found that an estimated minimum 1.25 years of extended life without needing active treatment to control myeloma was extremely desirable.

Myeloma Canada re-emphasized that cilta-cel is a therapy well understood by patients and caregivers but that it is also an expensive and resource-intensive therapy. The survey responses indicate that access to cilta-cel is currently difficult for patients in Canada, leading some patients to seek treatment outside the country.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

Unmet Needs

According to the clinical experts, the most important goal of treating patients with RRMM is to control disease with minimal toxicities, given that there are no curative therapies currently. The clinical experts indicated that patients with MM commonly experience drug resistance to each line of therapy, with progressively shorter durations of response. Additionally, the clinical experts highlighted that a treatment-free interval would be valuable to improve quality of life for patients, given that the current treatments for MM often require weekly or even twice-weekly injections, which is an inconvenience for patients. Therefore, the clinical experts stated that more treatment options are needed that work through novel pathways and can enhance and prolong treatment response with fewer side effects and improved convenience.

Place in Therapy

The clinical experts indicated that cilta-cel would be an additional option for the management of patients with MM whose disease is refractory to lenalidomide or who have been exposed to lenalidomide. The clinical experts confirmed that the proposed place in therapy (i.e., second to fourth line) is reflective of anticipated clinical practice in Canada. In general terms, for patients who are eligible for transplant, the clinical experts considered an mAB-based therapy (e.g., isatuximab-carfilzomib-dexamethasone [IsaKd] or daratumumab-bortezomib-dexamethasone [DVd]) as the preferred second-line treatment for patients who experience relapse after lenalidomide-bortezomib-dexamethasone (RVd) in the first line. Thus, cilta-cel may be preferred in the third line or later. However, cilta-cel could be a preferred second-line treatment for a small percentage (about 10%) of patients, such as those with higher-risk genetics or disease who received daratumumab-RVd in the first line. For patients who are not eligible for transplant, the clinical experts would promote cilta-cel as a second-line treatment but noted that, in clinical practice, most patients (80% to 90%) would receive daratumumab-lenalidomide-dexamethasone (DRd) as the first-line treatment, which would mean they are not eligible for cilta-cel in the second line. Generally, the clinical experts would not limit access to cilta-cel by mandating trying other treatments first, given that exposure to cilta-cel earlier in a patient’s disease course typically results in healthier and less exhausted T cells.

Patient Population

The clinical experts confirmed that the patients included in the CARTITUDE-4 trial are generally reflective of the patient population with MM in clinical practice in Canada. According to the clinical experts, given there is no companion test required and no established biomarker to identify those who may be most likely to benefit from cilta-cel, patients best suited for the treatment with cilta-cel would be identified through the professional judgment of physicians. Currently, as per feedback from the clinical experts, highly specialized testing, such as additional detailed genetic testing, is not widely available and not likely to become so in the near future.

Assessing the Response to Treatment

The clinical experts stated that standard clinical assessments of urine, blood, scans, and bone marrow are used to document response or relapse. These assessments include urine and serum protein electrophoresis and immune fixation, serum free light chains (FLCs), complete blood count, creatinine, calcium, and imaging (MRI, CT, PET-CT). The clinical experts mentioned that patient visits and blood assessments are usually done monthly initially and are then reduced to every 3 months for patients in remission and without symptoms. Imaging can be done with the onset of new symptoms or annually.

Discontinuing Treatment

As cilta-cel is a 1-time treatment, the clinical experts indicated that stopping treatment is not applicable.

Prescribing Considerations

The clinical experts stated that cilta-cel should be administered in qualified institutions that are capable of properly handling patient cells, including their acquisition, storage, and shipment. Additionally, the clinical experts indicated that specialized centres administering CAR T-cell therapy are required to have processes in place to manage acute toxicities occurring, usually, within the first 28 days postinfusion; examples include CRS (which, if present, requires intensive care to be available) and neurotoxicity (which, if present, requires neurologic care to be available). The management of patients with MM undergoing CAR T-cell therapy requires ongoing monitoring of immunity, revaccination, and immunoglobulin therapy administration according to the clinical experts.

Clinician Group Input

CDA-AMC received input from 2 clinician groups: 1 submission from the Ontario Health (OH)-Cancer Care Ontario (CCO) Hematology Cancer Drug Advisory Committee, which provides timely evidence-based clinical and health system guidance on drug-related issues in support of CCO’s mandate, including the Provincial Drug Reimbursement Programs and the Systemic Treatment Program, and 1 submission from the Canadian Myeloma Research Group (CMRG), a Canada-wide network of researchers aiming to develop better treatments to extend the life of myeloma patients, enhancing the quality of life for those living with myeloma and related disorders, and working to find a cure for these diseases and other plasma cell disorders. Both groups gathered information via teleconference.

OH-CCO indicated that cilta-cel is an option as second-line treatment for patients who are eligible for transplant or likely as third-line treatment for patients who are not eligible for transplant as they would get daratumumab in the first line. CMRG also emphasized that the availability of cilta-cel in the proposed setting would pertain primarily to patients who have had 2 prior lines of treatment; they may or not may not have already received an anti-CD38 mAB as well in the current treatment environment. CMRG further commented that the highest unmet need in myeloma continues to be adequate treatment for patients who have experienced disease progression despite exposure to an effective drug (for example, patients whose disease is triple-class refractory to an IMiD, PI, and anti-CD38 mAB). As combinations of these 3 major drug classes are increasingly used in first-line and second-line treatment, patients are now developing resistance to multiple drug classes much earlier in the disease course. OH-CCO also mentioned that patients who had been exposed to anti-CD38 mAB particularly had poor outcomes.

OH-CCO considered that improved response, quality of life, disease-related symptoms, PFS, and OS are important outcomes. CMRG highlighted that cilta-cel produces unprecedented rates of response that are deeper than standard regimens; specifically, the rates of complete response (CR) and stringent CR (sCR) are on the order of 70% to 75%, compared to 20% with standard therapy.

Both groups agreed that cilta-cel should be delivered at tertiary hospitals or transplant centres with expertise in cellular therapy with an intensive care unit familiar with patients with cancer who are immunosuppressed and an outpatient facility experienced in handling complex and urgent hematologic problems.

Drug Program Input

Input was obtained from the drug programs that participate in the reimbursement review process. The following were identified as key factors that could potentially affect the implementation of a recommendation for cilta-cel: relevant comparators, considerations for initiation of therapy, considerations for prescribing of therapy, generalizability, funding algorithm, care provision issues, and system and economic issues.

The clinical experts consulted for the purpose of this review provided advice on the potential implementation issues raised by the drug programs. Refer to Table 4 for more details.

Clinical Evidence

Systematic Review

Description of Studies

One pivotal trial (the CARTITUDE-4 trial) was included in the systematic review. The CARTITUDE-4 trial is an ongoing phase III, open-label, randomized, multicentre study to evaluate the efficacy and safety of cilta-cel compared to physician’s choice of standard-of-care (SOC) therapies of either pomalidomide-bortezomib-dexamethasone (PVd) or daratumumab-pomalidomide-dexamethasone (DPd) in patients with RRMM who have received 1 to 3 prior lines of therapy. The CARTITUDE-4 trial enrolled adults who had a documented diagnosis of MM according to International Myeloma Working Group (IMWG) diagnostic criteria, had received 1 to 3 prior lines of therapy, including a PI and an IMiD, and whose disease was refractory to lenalidomide per IMWG consensus guidelines. A total of 419 eligible patients were randomized at a 1:1 ratio to receive either cilta-cel (n = 208) or standard therapy with PVd or DPd (n = 211). Randomization was stratified by physician’s choice of PVd or DPd, international staging system (ISS) disease stage at screening (I, II, or III), and number of prior lines of therapy (1 versus 2 to 3). The median age of all study participants was 61.0 years, with a range of 27 years to 80 years. The demographic characteristics and disease history were balanced between treatment groups. At baseline, most of the participants were at ISS disease stage I (64.0%), had had 2 lines of therapy (40.9%), and had at least 1 high-risk cytogenetic abnormality (61.2%), with gain/amp(1q) being the most-reported abnormality (47.0%) in all patients. The primary objective of the study was to compare the efficacy of cilta-cel with SOC of either PVd or DPd in terms of PFS in patients with relapsed and lenalidomide-refractory MM. The primary end point was PFS according to a computerized algorithm per IMWG criteria, and secondary and other end points included CR or better rate, very good partial response (VGPR) or better rate, overall response rate (ORR), minimal residual disease (MRD) negativity rate, OS, duration of response (DOR), and HRQoL. The study was funded by Janssen and Legend Biotech.

Efficacy Results

Only those efficacy outcomes identified as important for this review are reported. Efficacy and safety data were evaluated at a planned interim analysis with data cut-off date of November 1, 2022.

Progression-Free Survival

In the interim analysis, 65 patients (31.3%) in the cilta-cel treatment group and 122 patients (57.8%) in the SOC group experienced an event. With a median follow-up of 15.8 months in the cilta-cel group and 15.3 months in the SOC group, the median PFS was not reached (95% confidence interval [CI], 22.8 months to not estimable) for the cilta-cel group and was 11.8 months (95% CI, 9.7 months to 13.8 months) for the SOC group. The Kaplan-Meier estimate of PFS probabilities decreased from 75.9% (95% CI, 69.4% to 81.1%) ██ █████ ████ ███ █████ ██ ██████ in the cilta-cel group and 48.6% (95% CI, 41.5% to 55.3%) to █████ ████ ███ ████ ██ ██████ in the SOC group from 12 to 24 months. The PFS results were consistent across all prespecified and additional sensitivity analyses and subgroup results.

CR or Better Rate

The CR or better rate was higher in the cilta-cel group than in the SOC group (73.1% versus 21.8% for cilta-cel versus SOC; odds ratio [OR] = 10.3; 95% CI, 6.5 to 16.4; P < 0.0001).

VGPR or Better Rate

A total of 169 patients (81.3%) in the cilta-cel group and 96 patients (45.5%) in the SOC group reported a VGPR or better (OR = 5.9; 95% CI, 3.7 to 9.4; nominal P < 0.0001).

Overall MRD Negativity Rate

A higher proportion of patients in the cilta-cel group than in the SOC group were reported to have negative overall MRD by next-generation sequencing in bone marrow (60.6% versus 15.6% for cilta-cel versus SOC; OR = 8.7; 95% CI, 5.4 to 13.9; P < 0.0001).

Overall Survival

With a median follow-up of 16.0 months for the cilta-cel group and 15.9 months for the SOC group, median OS was not reached in the cilta-cel group and was 26.7 months (95% CI, 22.5 months to not estimable) in the SOC group. The Kaplan-Meier estimate of OS probabilities decreased from 84.1% (95% CI, 78.4% to 88.4%) to █████ ████ ███ █████ ██ ██████ in the cilta-cel group and from 83.6% (95% CI, 77.8% to 88.0%) ██ █████ ████ ███ █████ ██ ██████ in the SOC group from 12 to 24 months.

Duration of Response

With a median follow-up of 13.7 months for the cilta-cel group and 14.3 months for the SOC group, the median DOR was not reached in the cilta-cel group and was 16.6 months (95% CI, 28.9 months to not estimable) in the SOC group. Among patients who had a partial response (PR) or better (176 versus 142 for cilta-cel versus SOC), 143 patients (81.3%) in the cilta-cel group and 80 patients (56.3%) in the SOC group were censored. The Kaplan-Meier estimate of event-free probabilities decreased from 84.7% (95% CI, 78.1% to 89.4%) to █████ ████ ███ █████ ██ ██████ in the cilta-cel group and from 63.0% (95% CI, 54.2% to 70.6%) to █████ ████ ███ █████ ██ ██████ in the SOC group from 12 to 24 months.

Time to Worsening of Symptoms in the Multiple Myeloma Symptom and Impact Questionnaire Total Symptom Score

The median time to a sustained worsening of MM symptoms was longer for the cilta-cel group (23.7 months) than for the SOC group (18.9 months), with a hazard ratio (HR) of 0.42 (95% CI, 0.26 to 0.68; nominal P = 0.0003). The Kaplan-Meier estimate of event-free probabilities decreased from 84.6% (95% CI, 77.7% to 89.6%) to 79.8% (95% CI, 69.6% to 86.9%) in the cilta-cel group and from 65.6% (95% CI, 55.2% to 74.2%) to 51.9% (95% CI, 34.5% to 66.8%) in the SOC group from 12 to 18 months.

Harms Results

All patients in both treatment groups reported at least 1 treatment-emergent adverse event (TEAE) in the interim analysis (data cut-off: November 1, 2022). The most commonly reported adverse events (i.e., reported by at least 20% of patients in either group) were blood and lymphatic system disorders, including neutropenia (89.9% versus 85.1% for cilta-cel versus SOC); immune system disorders (77.5% versus 8.2%); gastrointestinal disorders (74.0% versus 55.8%); thrombocytopenia (54.3% versus 31.3%); and anemia (54.3% versus 26.0%). Serious adverse events (SAEs) were reported among 44.2% of patients in cilta-cel group and 38.9% of patients in the SOC group. Infections and infestations (24.0% versus 24.5%), including COVID-19 pneumonia (5.8% versus 4.3%), was the most reported SAE. Withdrawals due to TEAEs were reported among | ████████ █████ in the cilta-cel group and ██ ████████ ███████ in the SOC group. The sponsor and/or the clinical experts identified notable harms as including CRS, neurotoxicity (including immune effector cell–associated neurotoxicity syndrome [ICANS]), B-cell aplasia, hypogammaglobulinemia, and immune suppression. CRS was reported for 76.1% patients in the cilta-cel group (134 of 176 patients), with the majority (52.8%) being grade 1. Only 2 patients (1.1%) experienced grade 3 CRS, and no grade 4 or 5 CRS was reported. In total, 36 patients (20.5%) from the cilta-cel group experienced CAR T-cell neurotoxicity, including ICANS in 8 patients (4.5%). Among the 8 patients with ICANS, 6 patients (3.4%) had grade 1 events and 2 patients (1.1%) had grade 2 events. Hypogammaglobulinemia was observed in 88 of 202 patients (42.3%) in the cilta-cel group and 13 of 202 patients (6.3%) in the SOC group, with 15 patients (7.2%) in the cilta-cel group and 1 patient (0.5%) in the SOC group experiencing grade 3 or 4 hypogammaglobulinemia. Immune suppression was observed in 186 patients (89.4%) in the cilta-cel group and 182 patients (87.5%) in the SOC group. No data for B-cell aplasia were reported.

Critical Appraisal

In the CARTITUDE-4 trial, at baseline, higher proportions of patients received concomitant antimicrobial and antiviral medications, normal human immunoglobulin, serotonin (5-HT3) antagonists, paracetamol, and enoxaparin in the cilta-cel group than in the SOC group, some of which might have had an impact on the frequency of reported adverse events in the cilta-cel group. Additionally, patients in the cilta-cel group reported more frequent concomitant use of interleukin inhibitors than patients in the SOC group. According to the clinical experts, interleukin inhibitors are immunosuppressants, which could decrease T-cell function, which may bias the efficacy results against cilta-cel. Fewer patients received subsequent anticancer treatment in the cilta-cel group than in the SOC group; the review team agreed with the clinical experts that this would bias the study’s subsequent OS results against cilta-cel. A higher proportion of patients in the cilta-cel group discontinued and did not receive the study treatment than in the SOC group (██████ ███ ████ for cilta-cel versus SOC). The review team noted that the differential imbalance in the baseline characteristics of the patients who discontinued treatment between the 2 groups could have been a source of attrition bias against the cilta-cel group.

As the CARTITUDE-4 trial is ongoing, results were only available from the interim analysis (data cut-off: November 1, 2022), and the median PFS and median OS had not been reached in the cilta-cel group at the time of the interim analysis. Although results from the sponsor-conducted subsequent OS analysis (data cut-off: December 13, 2023) indicated a trend favouring OS benefit for the cilta-cel group compared to the SOC group, the median OS was still not yet reached at this time. Moreover, the statistical testing of the subsequent OS analysis was not controlled for the overall type I error; therefore, the results were descriptive and should be considered as supportive data. Many of the outcomes used in the CARTITUDE-4 trial (PFS, OS, CR or better rate, VGPR or better rate, ORR, and DOR) were identified as clinically important by patients and/or clinicians. However, VGPR or better rate and DOR were not part of the statistical testing strategy and thus were not adjusted for multiple testing; therefore, the ability to draw conclusions from these data may be limited.

It is uncertain to what extent the observed OS, patient-reported HRQoL, and disease symptom results from the CARTITUDE-4 trial could be generalized to clinical practice in Canada considering the limited representativeness of the study population due to restrictive eligibility criteria and comparators. The eligibility criteria for the CARTITUDE-4 trial excluded a small group of patients (less than 5%) with symptomatic MM who did not have measurable disease and patients with an Eastern Cooperative Oncology Group performance status (ECOG PS) of 2. The clinical experts opined that those patients would not necessarily be excluded from eligibility for cilta-cel. The clinical experts noted that patients who have confirmed relapsed disease, even if nonsecretory, may still benefit from therapy with cilta-cel. Those with an ECOG PS of 2, especially if the poor performance score is due to myeloma disease burden, may also benefit. Careful consideration of overall health and ability to withstand acute toxicities such as CRS would be important. The comparators used in the trial (i.e., PVd and DPd) may not be exactly reflective of the current clinical practice in Canada, and there was no study site in Canada in the CARTITUDE-4 trial. However, comparable triplet regimens are used; thus, findings are relevant to clinical practice in Canada. Patient and clinician groups indicated that prolonging PFS and OS, delaying progression, maintaining HRQoL, and controlling the symptoms of the disease were critical treatment considerations.

GRADE Summary of Findings and Certainty of the Evidence

For the pivotal CARTITUDE-4 trial identified in the sponsor’s systematic review, Grading of Recommendations Assessment, Development and Evaluation (GRADE) was used to assess the certainty of the evidence for the outcomes considered most relevant to inform expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^12,13 Following the GRADE approach, evidence from randomized controlled trials (RCTs) started as high-certainty evidence and could be rated down for concerns related to study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias.

The selection of outcomes for GRADE assessment was based on the sponsor’s summary of clinical evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members: PFS, CR or better rate, VGPR or better rate, overall MRD negativity rate, OS, DOR, HRQoL, and SAEs. As per feedback from the clinical experts, ORR was not used in the GRADE assessment as it represents patients with any type of response, which may not be as informative as the CR or better rate or the VGPR or better rate in providing clinically relevant information as an efficacy outcome.

Table 2: Summary of Findings for Cilta-Cel Versus SOC for Patients With RRMM

CI = confidence interval; cilta-cel = ciltacabtagene autoleucel; CR = complete response; MID = minimal important difference; MRD = minimal residual disease; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire; NA = not available; NR = not reported; OR = odds ratio; PFS = progression-free survival; PR = partial response; RCT = randomized controlled trial; RRMM = relapsed or refractory multiple myeloma; sCR = stringent complete response; SAE = serious adverse event; SOC = standard of care; VGPR = very good partial response.

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

^aRated down 1 level for serious imprecision. There is no established between-group MID for PFS at 12 months, but the clinical experts considered that a 20% difference between groups in the probability of patients being alive and progression-free could be considered a threshold of clinical importance. The point estimate and the upper bound of the 95% CI for the between-group difference suggested a clinically important difference for cilta-cel vs. SOC based on a 20% threshold, while the lower bound of the 95% CI suggested no clinically important difference between the 2 groups. In the absence of available data for the between-group difference in PFS probabilities at 24 months, the judgment of imprecision was based on the point estimates per different groups using the null as the threshold. The clinical importance of the between-group difference was judged based on the input of the clinical experts and the observed trend of differences, which was consistent with that at 12 months.

^bThe estimates of between-group difference were not available. The sponsor indicated that the 24-month PFS, OS, and DOR data were immature given that the median follow-up of study duration was 15.9 months at the interim analysis.

^cImprecision was not rated down. There is no established MID, and the clinical experts could not provide a threshold of important difference, so the target of the certainty appraisal was any effect for the outcome.

^dThe statistical testing for VGPR or better rate was not adjusted for multiplicity in the CARTITUDE-4 trial and should be considered as supportive evidence.

^eRated down 1 level for serious imprecision. There is no established MID, and the clinical experts could not provide a threshold of important difference, so the target of the certainty appraisal was any effect for overall survival. At 12 months, the lower bound of the 95% CI for the between-group difference was below zero, while the upper bound was above zero, suggesting no clinically important difference between the 2 groups. At 24 months, given that the between-group difference in overall survival probabilities was not available due to the immaturity of the data, as indicated by the sponsor, the judgment of imprecision was based on the point estimates per different groups using the null as the threshold and the observed trend of differences, which was similar to that at 12 months.

^fImprecision was not rated down. There is no established MID, and the clinical experts could not provide a threshold of important difference, so the target of the certainty appraisal was any effect for the outcome. In the absence of available data for the between-group difference in the probability of remaining in response (CR, sCR, VGPR, or PR) at 24 months, the judgment of imprecision was based on the point estimates per different groups using the null as the threshold.

^gRated down 1 level for serious risk of bias. Consistently and notably higher proportions of patients in the cilta-cel group than in the SOC group received various concomitant therapies for the control of various clinical symptoms or disorders associated with the increased incidence of adverse events. Given that the frequency and/or severity of adverse events might significantly affect patients’ HRQoL, including daily functions, the imbalances in concomitant medications may bias the HRQoL results in favour of cilta-cel.

^hRated down 2 levels for very serious imprecision. There is no established MID, and the clinical experts suggested that 10% is the threshold of important difference in the proportion of patients with at least 1 SAE. The point estimate and lower bound of the 95% CI for the between-group difference suggested no clinically important difference between the groups; the upper bound of the 95% CI for difference between groups suggested a clinically important harm of cilta-cel.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial¹⁴ and sponsor-provided additional data.^15,16 Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Long-Term Extension Studies

No long-term extension studies were submitted for this review.

Indirect Comparisons

Description of Studies

Two reports of indirect treatment comparison (ITC) analyses were submitted by the sponsor. One ITC report was based on the inverse probability of treatment weighting (IPTW) analyses, in which individual patient data (IPD) from daratumumab trials including the CANDOR (carfilzomib-dexamethasone [Kd]), CASTOR (bortezomib-dexamethasone [Vd], DVd), and APOLLO (pomalidomide-dexamethasone [Pd]) trials were matched to the eligibility criteria of CARTITUDE-4 trial to inform the comparison to cilta-cel. Another ITC report presented unanchored matching-adjusted indirect comparison (MAIC) analyses, matching IPD from the cilta-cel group of the CARTITUDE-4 trial to the isatuximab-pomalidomide-dexamethasone (IsaPd) group of the ICARIA-MM trial and the selinexor-bortezomib-dexamethasone (SVd) group of the BOSTON trial. The comparative treatment effect on outcomes of interest was reported, including tumour response outcomes (ORR, VGPR or better, CR or better) and survival outcomes (PFS and OS). The base-case scenario for tumour response and PFS outcomes incorporated 4 variables in the IPTW analyses (refractory status, ISS disease stage, presence of plasmacytomas or extramedullary disease, time to progression on prior line) and 3 variables in the MAIC analyses (refractory status, cytogenetic risk, ISS disease stage). Assessment of OS was conducted via multivariable regression, with 14 prognostic variables used for adjustment.

In the IPTW analyses, the CARTITUDE-4 study data excluded patients with prior anti-CD38 mAB therapy, leading to a sample size of 155 patients for the comparative analyses. The comparator treatment populations consisted of the following cohorts: 44 patients treated with DVd (CASTOR trial), 46 patients treated with Vd (CASTOR trial), 46 patients treated with Kd (CANDOR trial), and 92 patients treated with Pd (APOLLO trial).

In the MAIC analysis, the CARTITUDE-4 study initially consisted of 208 patients who were treated with cilta-cel. The number of patients in the IsaPd and SVd cohorts was 154 and 53, respectively. Following MAIC adjustment, the cilta-cel effective sample size (ESS) was 26 for the comparison to IsaPd (ICARIA-MM trial) and 188 for the comparison to SVd (BOSTON trial).

Efficacy Results

Progression-Free Survival

IPTW-Based Analyses

The observed median PFS for cilta-cel was not reached. The observed median PFS for DVd, Vd, Kd, and Pd was 7.59 months (95% CI, 6.51 to 11.17 months), 4.93 months (95% CI, 3.98 to 6.57 months), 12.01 months (95% CI, 7.43 to 15.26 months), and 6.93 months (95% CI, 4.73 to 9.53 months), respectively. The median PFS using IPTW was 9.79 months (95% CI, 6.51 to 13.40 months) for DVd, 6.21 months (95% CI, 3.84 to 7.03 months) for Vd, 11.09 months (95% CI, 3.98 to 15.26 months) for Kd, and 8.34 months (95% CI, 2.14 to 9.26 months) for Pd.

Following IPTW adjustment, the conditional HR for PFS between cilta-cel and Kd was 0.27 (95% CI, 0.16 to 0.45), between cilta-cel and Pd was 0.19 (95% CI, 0.13 to 0.30), between cilta-cel and Vd was 0.11 (95% CI, 0.07 to 0.17), and between cilta-cel and DVd was 0.25 (95% CI, 0.15 to 0.41), all favouring cilta-cel.

Unanchored MAIC Analyses

The median adjusted PFS for cilta-cel was not reached. The median PFS for IsaPd and SVd was █████ ██████ ████ ███ ████ ██ ██████ and ████ ██████ ████ ███ ████ ██ ███████ respectively.

The MAIC-adjusted HR for PFS between treatment groups was 0.32 (95% CI, 0.15 to 0.70) in the cilta-cel versus IsaPd comparison and ████ ████ ███ ████ ██ █████ | in the cilta-cel versus SVd comparison.

Overall Survival

IPTW-Based Analyses

The observed median OS for cilta-cel and Kd was not reached. The observed median OS for DVd, Vd, and Pd was █████ ██████ ████ ███ █████ ██ ██████, █████ ██████ ████ ███ █████ ██ ███████ and █████ ██████ ████ ███ █████ ██ ███████ respectively.

Following adjustment, the conditional HR for OS between cilta-cel and Kd was ████ ████ ███ ████ ██ █████, between cilta-cel and Pd was ████ ████ ███ ████ ██ █████, between cilta-cel and Vd was ████ ████ ███ ████ ██ ████, and between cilta-cel and DVd was ████ ████ ███ ████ ██ ██████.

Unanchored MAIC Analyses

The median OS for cilta-cel was not reached. The median OS for IsaPd and SVd was ████ ██████ ████ ███ ████ ██ █████ and ████ ██████ ████ ███ ████ ██ ████ respectively. The adjusted HR for OS was ████ ████ ███ ████ ██ ████ in the cilta-cel versus IsaPd comparison and ████ ████ ███ ████ ██ █████ in the cilta-cel versus SVd comparison.

Overall Response Rate

IPTW-Based Analyses

The observed ORR in the treatment populations was 89.7% for cilta-cel, 76.1% for Kd, 42.4% for Pd, 54.4% for Vd, and 72.7% for DVd. The IPTW-estimated relative risk (RR) was 1.32 (95% CI, 0.99 to 1.74) for cilta-cel versus Kd, 2.00 (95% CI, 1.31 to 3.06) for cilta-cel versus Pd, 1.77 (95% CI, 1.19 to 2.65) for cilta-cel versus Vd, and 1.38 (95% CI, 0.86 to 2.20) for cilta-cel versus DVd.

Unanchored MAIC Analyses

Observed proportions in the treatment populations were 84.6% for cilta-cel, 60.4% for IsaPd, and █████ | for SVd. The MAIC-estimated RR wase 1.39 (95% CI, 1.19 to 1.63) for cilta-cel versus IsaPd, and ████ ████ ███ ████ ██ ████ for cilta-cel versus SVd.

CR or Better

IPTW-Based Analyses

The observed CR or better rate in the treatment populations was 78.1% for cilta-cel, 10.9% for Kd, 2.2% for Pd, 4.4% for Vd, and 11.4% for DVd. The IPTW-estimated RR was 6.48 (95% CI, 2.72 to 15.43) for cilta-cel versus Kd, 38.76 (95% CI, 8.55 to 175.8) for cilta-cel versus Pd, 15.60 (95% CI, 3.88 to 62.73) for cilta-cel versus Vd, and 9.36 (95% CI, 3.35 to 26.14) for cilta-cel versus DVd.

Unanchored MAIC Analyses

The observed CR or better rate in the treatment populations was 73.1% for cilta-cel, 4.5% for IsaPd, and ████ for SVd. The MAIC-estimated RR was 17.30 (95% CI, 8.29 to 36.11) for cilta-cel versus IsaPd and ████ ████ ███ ████ ██ ██████ for cilta-cel versus SVd.

VGPR or Better

IPTW-Based Analyses

The observed VGPR or better rate in the treatment populations was 85.2% for cilta-cel, 52.2% for Kd, 14.1% for Pd, 15.2% for Vd, and 40.9% for DVd. The IPTW-estimated RR was 1.81 (95% CI, 1.24 to 2.64) for cilta-cel versus Kd, 3.73 (95% CI, 1.52 to 9.15) for cilta-cel versus Pd, 5.13 (95% CI, 2.39 to 10.99) for cilta-cel versus Vd, and 2.51 (95% CI, 1.39 to 4.53) for cilta-cel versus DVd.

Unanchored MAIC Analyses

The observed VGPR or better rate in the treatment populations was 81.3% for cilta-cel, 31.8% for IsaPd, and █████ for SVd. The MAIC-estimated RR was 2.52 (95% CI, 1.95 to 3.25) for cilta-cel versus IsaPd and ████ ████ ███ ████ ██ █████ for cilta-cel versus SVd.

Harms Results

Sponsor-conducted ITCs did not evaluate the comparative safety of cilta-cel.

Critical Appraisal

The sponsor-conducted IPTW analyses demonstrated favourable benefits with cilta-cel relative to Kd, Pd, Vd, and DVd treatments, though important limitations were noted. Heterogeneity between the CARTITUDE-4 trial and the comparator trials was observed, both in terms of study eligibility criteria and baseline population characteristics. Reduced sample sizes were generated and used in the analyses, after matching and adjustment methods. Certain prognostic factors, such as cytogenetic risk and type of previous treatment regimen, were unavailable for the adjustment in the IPTW analyses. Further uncertainty is associated with the possibility of unknown, unmeasured, or unmeasurable confounders, which cannot be accounted for with propensity score methods. Regarding the assessment of survival outcomes, median PFS and OS were not reached for cilta-cel and there was evidence of a possible violation of proportional hazards assumptions for certain comparisons (i.e., possible visual violation observed for the DVd, Kd, and Pd comparisons [PFS outcome] and the Kd comparison [OS outcome]; statistical violation, based on the Grambsch-Therneau test, observed for the Vd comparison [PFS outcome]). Input from the clinical expert suggested that certain important treatments of interest for clinical practice in Canada (IsaKd) were missing in the ITC analyses. Moreover, comparative safety and HRQoL were not evaluated, despite being considered important outcomes for patients with MM. Considering all the above, it is likely that the IPTW estimates are subject to an unknown amount and direction of bias.

Limitations of the sponsor-conducted unanchored MAIC included restrictions in ESSs for cilta-cel, following MAIC adjustments, and notable heterogeneity in prognostic and effect-modifying factors across the individual studies. The exploration of between-study differences was further limited by missing information on patient characteristics across the trials. Generalizability issues are associated with diverse eligibility criteria between the comparator and cilta-cel cohorts, mainly the inclusion of patients with an ECOG PS of 0,1, or 2 in the BOSTON trial and the inclusion of patients with at least 2 previous lines of treatment in the ICARIA-MM trial. Thus, concerns remain that not all prognostic and effect-modifying factors were accounted for in the unanchored comparisons, leading to challenges for interpretation and high uncertainty in the MAIC findings.

Studies Addressing Gaps in the Evidence From the Systematic Review

No studies addressing gaps in the systematic review evidence were submitted for this review.

Conclusions

Evidence derived from an ongoing trial demonstrated that the infusion of cilta-cel, compared with SOC (i.e., PVd or DPd), has shown a clinically significant benefit in terms of PFS in patients with RRMM who have received 1 to 3 prior lines of therapy and whose disease is refractory to lenalidomide followed by at least 1 bridging therapy cycle of either PVd or DPd. The treatment benefit with cilta-cel was also consistently presented in terms of CR, sCR, and MRD-negative status. The OS benefit was uncertain based on the submitted evidence due to immaturity of the data, as were the reported treatment effects on PFS and DOR. Results relating to patients’ HRQoL, as measured by a disease-specific quality-of-life instrument, the MySIm-Q total symptom score, were prone to bias due to increased use of concomitant therapies to control side effects, which would have positively impacted the quality of life of patients in the cilta-cel group.

There is low-certainty evidence that cilta-cel, when compared with SOC, may result in little to no difference in the percentage of patients who experience SAEs. Overall, no new safety signals were identified in the CARTITUDE-4 trial; the observed safety profile of cilta-cel is aligned with clinical practice as per feedback from the clinical experts.

In the ITCs comparing cilta-cel to various currently available therapies (i.e., Kd, Pd, Vd, DVd, IsaPd, and SVd), cilta-cel demonstrated statistically significant improvements in terms of PFS (for all comparisons) and OS (for the comparisons versus Pd, Vd, DVd, and SVd). No statistically significant differences were shown in terms of OS in the comparisons of cilta-cel versus Kd and IsaPd. However, the comparative evidence derived from ITC was associated with notable limitations, including incomplete adjustment of important effect modifiers and concern of restricted generalizability to the clinical setting in Canada.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of cilta-cel (Carvykti) cell suspension, administered via IV infusion at a dose of 0.5 × 10⁶ to 1.0 × 10⁶ CAR-positive viable T cells per kilogram of body weight, with a maximum of 1 × 10⁸ CAR-positive viable T cells, in the treatment of adult patients with MM who have received 1 to 3 prior lines of therapy, including a PI and an IMiD, and whose disease is refractory to lenalidomide.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CDA-AMC review team.

MM is a hematological malignancy characterized by clonal proliferation of malignant plasma cells (B-cells) driven by an oncologic event and consequent overproduction of the abnormal immunoglobulin M protein.¹ The accumulation of malignant plasma cells within the bone marrow and of monoclonal protein throughout the body leads to signs, symptoms, and complications that are characteristic of MM (e.g., Carbapenem-resistant Acinetobacter baumannii complications: hypercalcemia, renal failure, anemia, or lytic bone lesions).¹⁸

The disease has a highly heterogeneous presentation and clinical course. Almost all cases of MM are preceded by the premalignant, asymptomatic state of monoclonal gammopathy of undetermined significance (MGUS) or an intermediate stage of smouldering MM. Patients with MGUS progress to MM or related malignancy at a rate of 1% per year, while smouldering MM is associated with a much faster rate of progression than MGUS.¹⁸

MM has historically been associated with the lowest patient HRQoL of all hematological cancers,¹⁵ especially with increased duration of disease, disease progression, and consequently increasing lines of therapy.^19-24

It was estimated that 3,900 people in Canada would be diagnosed with MM and 1,700 would die from MM in 2023.³ Despite the improvement in the life expectancy of patients with MM over the last 2 decades, since the introduction of targeted therapies for MM, OS remains poor,^25-27 with only about half of patients remaining alive at 5 years after diagnosis.²⁸ As MM progresses, patient outcomes worsen with each subsequent line of therapy.^5-8

According to the Canadian Cancer Statistics 2022 report produced by the Canadian Cancer Society, Statistics Canada, and the Public Health Agency of Canada, in collaboration with the provincial and territorial cancer registries, the 2-year, 5-year, and 25-year prevalence of MM was 4,960 cases, 9,570 cases, and 15,030 cases, respectively, as of January 1, 2018, translating to a prevalence between 0.01% and 0.04%.⁴ According to the same report, the number of newly diagnosed cases of MM in 2022 was estimated as 4,000 cases, while the predicted 5-year and 10-year net survival was 50% (95% CI, 49% to 52%) and 30% (95% CI, 28% to 32%), respectively.⁴ Based on the reported MM prevalence in 2018 and the growing projected annual incidence rate, combined with a predicted 5-year survival rate, the projected prevalence of MM is estimated to be approximately 17,568 in Canada (excluding Quebec) in 2025.⁴

Diagnosis of MM usually begins with a visit to a primary care physician, occurring when a blood test for another condition is ordered, or if MM is suspected based on symptoms.^29,30 As other conditions can cause similar symptoms to MM, it is important for health care professionals to rule out other health conditions before diagnosing MM.²⁹ The IMWG revised the diagnostic criteria for MM in 2016 due to the remarkable progress made in the diagnosis and treatment of MM, including novel treatment options and advances in laboratory and imaging techniques.^18,31 The key revisions included the presence of 1 or more myeloma-defining events in addition to evidence of either 10% or more clonal plasma cell on bone marrow determined by an examination of a biopsy-proven plasmacytoma and the addition of 3 specific biomarkers: clonal plasma cells greater than 60% of plasma cells, serum FLC ratio greater than 100, and more than 1 focal lesion found via MRI.¹⁸

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CDA-AMC review team.

The treatment landscape for MM has changed significantly in recent years, with the emergence of new therapies in newly diagnosed and RRMM settings.³² According to clinical experts and clinician groups, the main treatment goals for patients with RRMM are to prolong survival, improve symptoms by extending remissions, minimize toxicities, and improve HRQoL.

According to the clinical experts and clinician groups consulted for the purpose of this review, in the first line of therapy, patients with MM are largely divided into those who are eligible for transplant and those who are not eligible for transplant. Patients who are eligible for transplant typically undergo induction therapy with either cyclophosphamide-bortezomib-dexamethasone or RVd, followed by autologous stem cell transplant and maintenance therapy with lenalidomide monotherapy until disease progression. Patients who are not candidates for autologous stem cell transplant receive 1 of several regimens, which include lenalidomide-based triplet therapy consisting of PIs, IMiDs, or mABs. Based on the inputs from the clinical experts and clinician groups, for patients who are not eligible for transplant, DRd is used preferentially across Canada; lenalidomide-dexamethasone and RVd are used less often than DRd.

In the second-line setting, for patients who are eligible for transplant and have experienced disease progression on lenalidomide, the clinical experts and clinician groups noted that, in the absence of prior exposure to daratumumab (an anti-CD38 mAB), IsaKd was stated to be the most commonly used treatment, followed by DVd; other potential treatment options include Kd, PVd, and SVd. For patients who are not eligible for transplant and whose disease is refractory to lenalidomide, potential combinations include Kd, SVd, or PVd. Most of these patients will have received a first-line anti-CD38 mAB combination, and thus combinations with anti-CD38 mABs are not used in the second line. In this situation, SVd may be a preferred second-line option because Kd could be used after disease progression, but the reverse is not true. For a small group of patients who are not eligible for transplant and who have not yet received an mAB as part of their first-line therapy and have experienced disease progression on lenalidomide, DVd or IsaKd are preferred, although PVd or SVd are also options. In the third-line setting, the clinical experts and clinician groups mentioned that IsaPd, Kd, Pd, and SVd are available treatment options. In the fourth-line setting, according to the clinical expert and clinician groups, treatment options — which include Kd, Pd, and SVd — have been extremely limited while cilta-cel has been under consideration for negotiation by the pan-Canadian Pharmaceutical Alliance. Typically, treatment options for later lines of therapy depend on a patient’s treatment history and response to previous lines of therapy. Thus, patients who become resistant to a previous regimen will be treated with a different regimen in later lines of therapy (with the exception of dexamethasone and cyclophosphamide, which can be used in multiple combination regimens).³³

Drug Under Review

Key characteristics of cilta-cel are summarized in Table 3, with other treatments available for the treatment of adult patients with MM who have received 1 to 3 prior lines of therapy, including a PI and an IMiD, and whose disease is refractory to lenalidomide.

Table 3: Key Characteristics of Cilta-Cel, Daratumumab, Selinexor, Carfilzomib, and Pomalidomide

BCMA = B-cell maturation antigen; ADCC = antibody-dependent cellular cytotoxicity; ADCP = antibody-dependent cellular phagocytosis; CAR = chimeric antigen receptor; CDC = complement-dependent cytotoxicity; cilta-cel = ciltacabtagene autoleucel; DRESS = drug rash with eosinophilia and systemic symptoms; DVd = daratumumab-bortezomib-dexamethasone; DVT = deep venous thrombosis; ICANS = immune effector cell–associated neurotoxicity syndrome; IMiD = immunomodulatory drug; Kd = carfilzomib-dexamethasone; mAB = monoclonal antibody; MM = multiple myeloma; PI = proteasome inhibitor; PML = progressive multifocal leukoencephalopathy; SC = subcutaneous; SJS = Stevens-Johnson syndrome; TEN = toxic epidermal necrolysis; TLS = tumour lysis syndrome.

^aHealth Canada–approved indication.

Sources: Product monographs for Carvykti,³⁴ Darzalex,³⁵ Xpovio,³⁶ Pomalyst,³⁷ Sarclisa,³⁸ and Kypolis.³⁹

Perspectives of Patients, Clinicians, and Drug Programs

Patient Group Input

This section was prepared by the CDA-AMC review team based on the input provided by patient groups. The full original patient input(s) have been included in the Patient, Clinician, and Drug Program Input section of this report.

CDA-AMC received 1 patient group submission, from Myeloma Canada. Myeloma Canada is a patient advocacy group that supports patients diagnosed with MM living in Canada.

Myeloma Canada collected data from patients with RRMM who had received 1 to 3 prior lines of therapy and whose disease was refractory to lenalidomide or who had experience with a CAR T-cell therapy, or from the caregivers of such patients, through a survey across Canada and internationally, via email and social media from April 5, 2024, to April 24, 2024. There were 53 eligible respondents; 51 lived in Canada (Alberta [3], British Columbia [12], Newfoundland and Labrador [2], Ontario [29], Quebec [5]), and 2 lived in France. There were 2 subsets of survey respondents: 1 subset comprised 37 patients or caregivers who met the criteria for the indication under review, and the other comprised 16 respondents who had CAR T-cell therapy experience, of which 8 patients or caregivers had experience with cilta-cel and 8 patients or caregivers had experience with a different CAR T-cell therapy.

In terms of MM disease complications, infections were considered the most important aspect to control, followed by kidney problems. Patients and/or caregivers also reported that MM had various impacts on their quality of life, such as limiting their ability to travel and their pursuit of life goals or accomplishments. Most patients and caregivers identified a need for effective MM treatment options, with manageable side effects and minimal impact on quality of life. Of the 37 patients or caregivers who met the criteria of the indication under review, 22 reported receiving 3 lines of therapy and 2 reported treatment with B-cell maturation antigen–targeted therapy. The experiences shared by patients or caregivers who received CAR T-cell therapy were generally positive. Of the 8 respondents who received cilta-cel, 5 rated the treatment extremely effective and the side effects extremely tolerable. CRS was perceived to be the most concerning side effect by patients who met the criteria for the indication under review. However, it was considered bearable for respondents who had received cilta-cel. Twenty-eight respondents out of the 37 found that an estimated minimum 1.25 years of extended life without needing active treatment to control myeloma was extremely desirable.

Myeloma Canada re-emphasized that cilta-cel is a therapy well understood by patients and caregivers but that it is also an expensive and resource-intensive therapy. The survey responses indicate that access to cilta-cel is currently difficult for patients in Canada, leading some patients to seek treatment outside the country.

Clinician Input

Input From Clinical Experts

All 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 (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of MM.

Unmet Needs

According to the clinical experts, the most important goal of treating patients with RRMM is to control disease with minimal toxicities, given there are no curative therapies currently. The clinical experts indicated that patients with MM commonly experience drug resistance to each line of therapy, with progressively shorter durations of response. Additionally, the clinical experts highlighted that a treatment-free interval would be valuable to improve quality of life for patients, given that the current treatments for MM often require injections weekly or sometimes even twice a week, which is an inconvenience for patients. Therefore, the clinical experts stated that more treatment options are needed that work through novel pathways and can enhance and prolong responses with fewer side effects and improved convenience.

Place in Therapy

The clinical experts indicated that cilta-cel would be an additional option for the management of patients with MM whose disease is refractory to lenalidomide or who have been exposed to lenalidomide. The clinical experts confirmed that the proposed place in therapy (i.e., second to fourth line) is reflective of clinical practice in Canada. In general terms, for patients who are eligible for transplant, the clinical experts considered an mAB-based therapy (e.g., IsaKd or DVd) as the preferred second-line treatment for patients who experience relapse after RVd in the first line. Thus, cilta-cel may be preferred in the third line or later. However, cilta-cel could be the preferred second-line treatment for a small percentage (about 10%) of patients, such as those with higher-risk genetics or disease who received daratumumab-RVd in the first line. The clinical experts felt that a trial to compare efficacy and toxicities between anti-CD38 mAB-based second-line therapy or second-line therapy with cilta-cel would provide important data to inform this decision. For patients who are not eligible for transplant, the clinical experts would promote cilta-cel as a second-line treatment but noted that, in clinical practice, most patients (80% to 90%) would receive DRd as the first-line treatment, which would mean they are not eligible for cilta-cel in the second line. Generally, the clinical experts would not limit access to cilta-cel by mandating trying other treatments first, given that exposure to cilta-cel earlier in a patient’s disease course typically results in healthier and less exhausted T cells.

Patient Population

The clinical experts confirmed that the patients included in the CARTITUDE-4 trial are generally reflective of the patient population with MM in clinical practice in Canada. According to the clinical experts, given there is no companion test required and no established biomarker to identify those who may be most likely to benefit from cilta-cel, the patients best suited for the treatment with cilta-cel would be identified by the physicians. Currently, as per feedback from the clinical experts, highly specialized testing, such as additional detailed genetic testing, is not widely available and not likely to become so in the near future.

Assessing the Response to Treatment

The clinical experts stated that standard clinical assessments of urine, blood, scans, and bone marrow are used to document response or relapse. These assessments include development of urine and serum protein electrophoresis and immune fixation, serum FLCs, complete blood count, creatinine, calcium, and imaging (MRI, CT, PET-CT). The clinical experts mentioned that patient visits and blood assessments are usually done monthly initially and are then reduced to every 3 months for patients in remission and without symptoms. Imaging can be done with the onset of new symptoms or annually.

Discontinuing Treatment

As cilta-cel is a 1-time treatment, the clinical experts indicated that stopping treatment is not applicable.

Prescribing Considerations

The clinical experts stated that cilta-cel should be administered in qualified institutions that are capable of properly handling patient cells, including their acquisition, storage, and shipment. Additionally, the clinical experts indicated that specialized centres administering CAR T-cell therapy are required to have processes in place to manage acute toxicities occurring, usually, within the first 28 days postinfusion; examples include CRS (which, if present, requires intensive care to be available) and neurotoxicity (which, if present, requires neurologic care to be available). The management of patients with MM undergoing CAR T-cell therapy requires ongoing monitoring of immunity, revaccination, and immunoglobulin therapy administration according to the clinical experts.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups. The full original clinician group input(s) received have been included in the Patient, Clinician, and Drug Program Input section of this report.

CDA-AMC received input from 2 clinician groups: 1 submission from the OH-CCO Hematology Cancer Drug Advisory Committee, which provides timely evidence-based clinical and health system guidance on drug-related issues in support of CCO’s mandate, including the Provincial Drug Reimbursement Programs and the Systemic Treatment Program, and 1 submission from CMRG, a Canada-wide network of researchers aiming to develop better treatments to extend the life of myeloma patients, enhancing the quality of life for those living with myeloma and related disorders, and working to find a cure for these diseases and other plasma cell disorders. Both groups gathered information via teleconference.

OH-CCO indicated that cilta-cel is an option as second-line treatment for patients who are eligible for transplant or likely as third-line treatment for patients who are not eligible for transplant as they would get daratumumab in the first line. CMRG also emphasized that the availability of cilta-cel in the proposed setting would pertain primarily to patients who have had 2 prior lines of treatment; they may or not may not have already received an anti-CD38 mAB as well in the current treatment environment. CMRG further commented that the highest unmet need in myeloma continues to be adequate treatment for patients who have experienced disease progression despite exposure to effective drugs (for example, patients whose disease is triple-class refractory to an IMiD, PI, and anti-CD38 mAB. Combinations of these 3 major drug classes are increasingly used in first-line and second-line treatment, and patients are now developing resistance to multiple drug classes much earlier in the disease course. OH-CCO also mentioned that patients who had been exposed to anti-CD38 mAB particularly had poor outcomes.

OH-CCO considered that improved response, quality of life, disease-related symptoms, PFS, and OS are important outcomes. CMRG highlighted that cilta-cel produces unprecedented rates of response that are deeper than standard regimens; specifically, the rates of sCR and CR are on the order of 70% to 75%, compared to 20% with standard therapy.

Both groups agreed that cilta-cel should be delivered at tertiary hospitals or transplant centres with expertise in cellular therapy with an intensive care unit familiar with patients with cancer who are immunosuppressed and an outpatient facility experienced in handling complex and urgent hematologic problems.

Drug Program Input

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

Table 4: Summary of Drug Plan Input and Clinical Experts’ Response

BCMA = B-cell maturation antigen; CAR = chimeric antigen receptor; cilta-cel = ciltacabtagene autoleucel; CRS = cytokine release syndrome; DPd = daratumumab-pomalidomide-dexamethasone; DRd = daratumumab-lenalidomide-dexamethasone; DVd = daratumumab-bortezomib-dexamethasone; ECOG PS = Eastern Cooperative Oncology Group performance status; ICANS = immune effector cell–associated neurotoxicity syndrome; ICU = intensive care unit; IsaKd = isatuximab-carfilzomib-dexamethasone; IsaPd = isatuximab-pomalidomide-dexamethasone; Kd = carfilzomib-dexamethasone; KRd = carfilzomib-lenalidomide-dexamethasone; NA = not applicable; PCd = pomalidomide-cyclophosphamide-dexamethasone; Pd = pomalidomide-dexamethasone; pERC = CADTH pan-Canadian Oncology Review Expert Review Committee; PI = proteasome inhibitor; PVd = pomalidomide-bortezomib-dexamethasone; SVd = selinexor-bortezomib-dexamethasone.

Clinical Evidence

The objective of this Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of cilta-cel, cell suspension in infusion bag, 0.5 × 10⁶ to 1.0 × 10⁶ CAR-positive viable T-cells per kilogram of body weight, with a maximum of 1 × 10⁸ CAR-positive viable T cells, for IV infusion in the treatment of RRMM in patients who have received 1 to 3 prior lines of therapy, including a PI and an IMiD, and whose disease is refractory to lenalidomide The focus will be placed on comparing cilta-cel to SOC and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of cilta-cel is presented in 2 sections, with critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes the pivotal study that was selected according to the sponsor’s systematic review protocol. The assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes sponsor-submitted ITCs.

Included Studies

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

• 1 phase III, open-label, randomized, active-control RCT (the CARTITUDE-4 trial) identified in the systematic review

• 2 ITC reports with IPTW analyses for 4 comparators and MAIC analyses for 2 comparators.

Systematic Review

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CDA-AMC review team.

Description of Studies

The characteristics of the included studies are summarized in Table 5.

Table 5: Details of Studies Included in the Systematic Review

BCMA = B-cell maturation antigen; CAR = chimeric antigen receptor; cilta-cel = ciltacabtagene autoleucel; CR = complete response; DPd = daratumumab-pomalidomide-dexamethasone; ECOG = Eastern Cooperative Oncology Group; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FLC = FLC; IMiD = immunomodulatory drug; IMWG = International Myeloma Working Group; MM = multiple myeloma; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PI = proteasome inhibitor; PR = partial response; PVd = pomalidomide-bortezomib-dexamethasone; RCT = randomized controlled trial; SC = subcutaneous; sCR = stringent complete response; SOC = standard of care; VGPR = very good partial response.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

One pivotal trial (the CARTITUDE-4 trial) was included in the systematic review. The CARTITUDE-4 trial is a phase III, open-label, randomized, multicentre study to evaluate the efficacy and safety of cilta-cel compared to physician’s choice of SOC therapies (PVd or DPd) in patients with RRMM who have received 1 to 3 prior lines of therapy. The primary objective of the study was to compare the efficacy of cilta-cel with SOC (PVd or DPd) in terms of PFS in patients with relapsed and lenalidomide-refractory MM. The primary end point was PFS, and secondary and other outcomes included CR or better rate, VGPR or better rate, ORR, MRD negativity rate, OS, DOR, and HRQoL. The study was funded by Janssen and Legend Biotech. The CARTITUDE-4 trial enrolled adults who had a documented diagnosis of MM according to IMWG diagnostic criteria, had received 1 to 3 prior lines of therapy, including a PI and an IMiD, and whose disease was refractory to lenalidomide per IMWG consensus guidelines.

The CARTITUDE-4 trial is an ongoing trial. Patient enrolment started on June 30, 2020. One interim analysis and 1 final analysis were planned. The interim analysis was performed when approximately 188 PFS events (75% of the total planned PFS events) had been observed. The final analysis is expected after the accumulation of approximately 250 PFS events, which has not yet occurred at the time of this report. Therefore, this report is focused on the results from the interim analysis. The data cut-off date was November 1, 2022, for the interim analysis. A total of 419 patients were randomized at a 1:1 ratio to receive either cilta-cel (n = 208) or standard therapy with PVd or DPd (n = 211). Randomization was stratified by physician’s choice of PVd or DPd, ISS disease stage at screening (I,. II, or III), and number of prior lines of therapy (1 versus 2 to 3). These patients were recruited in 81 centres across 16 countries in Europe (Belgium, Denmark, France, Germany, Greece, Italy, Netherlands, Poland, Spain, Sweden, and UK), North America (US), and other regions (Australia, Israel, Japan, and Republic of Korea). The study had no sites in Canada.

The study consists of 3 phases: screening, treatment, and follow-up (Figure 1). Before the screening phase, the investigator determined if the patient would be treated with PVd or DPd as standard therapy based on the patient’s prior exposure to antimyeloma therapies. In the screening phase, patients provided written consent for study participation and were screened for eligibility within 28 days before randomization. After screening, patients in the cilta-cel group underwent apheresis, received a conditioning treatment regimen of cyclophosphamide and fludarabine, and then received the cilta-cel infusion, which was administered 5 to 7 days after the start of the conditioning regimen. Patients were monitored for safety and efficacy during the first 112 days after cilta-cel administration (postinfusion follow-up). During the posttreatment follow-up (starting on day 112), patients continued to be monitored for efficacy until confirmed progressive disease, death, or withdrawal of consent. Patients who were unable to receive PVd or DPd standard therapy or who were unable to be apheresed or to receive bridging therapy, conditioning regimen, or cilta-cel infusion were followed until confirmed progressive disease, start of a new antimyeloma therapy, withdrawal of consent, or end of study, whichever occurred first. In the SOC group, patients received PVd or DPd until confirmed progressive disease, death, intolerable toxicity, withdrawal of consent, or end of study. For both groups, after confirmed progressive disease, patients were followed up for survival status, subsequent antimyeloma therapies, and the occurrence of second primary malignancies every 16 weeks until the end of the study. At the end of the study, all patients who received cilta-cel will continue to be monitored for long-term safety under a separate study for up to 15 years after cilta-cel administration. Adverse events were reported until 30 days after the last dose of bridging therapy or until day 112 postinfusion of cilta-cel (whichever is later). This reporting continued regardless of whether progressive disease occurred during bridging therapy or before day 112, or if a subsequent antimyeloma therapy is started before day 112 for the cilta-cel group. For the SOC groups, adverse events were reported until 30 days after the last dose of any study treatment or until the start of subsequent antimyeloma therapy (whichever occurred earlier). Adverse events that were considered to be related to the study drug but occurred after the adverse event reporting period were reported until the end of the study.

Figure 1: CARTITUDE-4 Study Schematic

CAR-T = chimeric antigen receptor T-cell therapy; DPd = daratumumab-pomalidomide-dexamethasone; PD = progressive disease; PVd = pomalidomide-bortezomib- dexamethasone; SPM = secondary primary malignancy.

Source: Statistical analysis plan for the CARTITUDE-4 trial.⁴¹

Populations

Inclusion and Exclusion Criteria

Eligible patients were required to be aged 18 years or older, have a documented diagnosis of MM according to IMWG diagnostic criteria, and have an ECOG PS of 0 or 1. Other key eligibility criteria included the following: measurable disease at screening; received 1 to 3 prior lines of therapy, including a PI and an IMiD; documented evidence of progressive disease by IMWG criteria based on investigator’s determination on or within 6 months of their last regimen; and have disease that is refractory to lenalidomide per IMWG consensus guidelines. Patients were excluded from the CARTITUDE-4 trial if they had received prior CAR T-cell therapy or B-cell maturation antigen–targeted treatment.

Interventions

Patients in the cilta-cel group underwent apheresis, followed by at least 1 bridging therapy cycle of either PVd or DPd (with the number of cycles based on clinical status and cilta-cel manufacturing time) and lymphodepletion with IV 300 mg of cyclophosphamide per square metre of body surface area and IV 30 mg of fludarabine per square metre of body surface area daily for 3 days. Five to 7 days after the initiation of lymphodepletion, a single cilta-cel infusion (target dose, 0.75 × 10⁶ CAR-positive viable T cells per kilogram of body weight) was administered. Patients in the cilta-cel group who had confirmed disease progression during bridging therapy or lymphodepletion were assessed as having a progression event and could receive cilta-cel as subsequent therapy at the investigator’s discretion.

In the SOC group, DPd was administered in 28-day cycles, and PVd in 21-day cycles, until disease progression, death, intolerable toxicity, withdrawal of consent, or the end of the study. The treatment schedule for PVd was as follows:

• For cycles 1 and 2:

  • Daratumumab subcutaneously: 1,800 mg (co-formulated with recombinant human hyaluronidase PH20) weekly, on days 1, 8, 15, and 22

  • Pomalidomide orally: 4 mg/day on days 1 to 21

  • Dexamethasone orally or IV: 40 mg weekly, on days 1, 8, 15, and 22, or could be split, with 20 mg given on days 1, 2, 8, 9, 15, 16, 22, and 23

• For cycles 3 to 6:

  • Daratumumab subcutaneously: 1,800 mg (co-formulated with recombinant human hyaluronidase PH20) every 2 weeks, on days 1 and 15

  • Pomalidomide orally: 4 mg/day on days 1 to 21

  • Dexamethasone orally or IV: 40 mg weekly (could be split over 2 days)

• For cycle 7 and onward:

  • Daratumumab subcutaneously: 1,800 mg (co-formulated with recombinant human hyaluronidase PH20) every 4 weeks on day 1

  • Pomalidomide orally: 4 mg/day on days 1 to 21

  • Dexamethasone orally or IV: 40 mg weekly (could be split over 2 days)

For all DPd cycles, on days of daratumumab administration, dexamethasone was given 1 to 3 hours before daratumumab.

The treatment schedule for PVd was as follows:

• For cycles 1 to 8:

  • Pomalidomide orally: 4 mg on days 1 to 14

  • Bortezomib subcutaneously: 1.3 mg/m² on days 1, 4, 8, and 11

  • Dexamethasone orally: 20 mg/day on days 1, 2, 4, 5, 8, 9, 11, and 12

• For cycle 9 and onward:

  • Pomalidomide orally: 4 mg on days 1 to 14

  • Bortezomib subcutaneously: 1.3 mg/m² on days 1 and 8

  • Dexamethasone orally: 20 mg/day on days 1, 2, 4, 5, 8, 9, 11, and 12

Crossover from the SOC group to the cilta-cel group was not permitted if patients had disease progression after standard treatment.

Throughout the study, investigators could prescribe concomitant medications or treatments deemed necessary to provide adequate supportive care, such as standard supportive care therapies (e.g., antiemetics, antidiarrheals, and anticholinergics), bisphosphonates, hematopoietic growth factor support and transfusions, antibiotics or anti-infective drugs, and chemotherapy drugs used to treat CAR T-cell–related toxicity. Unless prespecified in the study protocol, any chemotherapy, anticancer immunotherapy (other than cilta-cel), experimental therapies, other immunosuppressant drugs, or posttreatment medications to treat an adverse event were prohibited. Orthopedic surgery and radiotherapy were generally prohibited but could be allowed in the absence of disease progression. Additionally, RANKL inhibitors and pegylated myeloid growth factors were prohibited in the cilta-cel group.

Protocol Amendment

In the CARTITUDE-4 trial, 4 protocol amendments were reported. Protocol amendment 1, made March 20, 2020, removed eligibility for patients with any history of Parkinson disease or other neurodegenerative disorder from study enrolment. Protocol amendment 2, made July 2, 2021, added eligibility for patients with a history of autoimmune disease within 2 years and patients whose serum M protein level was between 0.5 g/dL and 1.0 g/dL. Protocol amendment 3, made June 14, 2022, informed investigators that patients receiving cilta-cel are possibly at higher risk of severe and/or fatal outcomes from COVID-19 infection than patients who are receiving SOC therapy and provided additional guidance for prevention and mitigation. Protocol amendment 4, made August 18, 2022, changed the number of PFS events required to trigger the interim analysis. Per health authority request, the interim analysis would take place after approximately 75% of the total PFS events had been observed.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 6, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the sponsor’s summary of clinical evidence as well as any outcomes identified as important to this review according to the clinical experts consulted for the purpose of this review and input from patient and clinician groups and public drug plans. Using the same considerations, the review team selected the end points that were considered most relevant to inform the expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing the expert committee deliberations were also assessed using GRADE.

Table 6: Outcomes Summarized From the CARTITUDE-4 Study Included in the Systematic Review

CR = complete response; MRD = minimal residual disease; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire; NR = not reported; sCR = stringent complete response; VGPR = very good partial response.

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., hierarchal testing).

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Efficacy Outcomes

Progression-Free Survival

PFS was the primary end point for the CARTITUDE-4 trial and was defined as the time from randomization to the date of first documented evidence of confirmed disease progression (based on a validated computerized algorithm) or death due to disease progression, whichever occurred first. The initiation of subsequent antimyeloma therapy was considered an intercurrent event; patients who started subsequent antimyeloma therapies for MM without disease progression were censored at the last disease assessment before the start of subsequent therapies. PFS was considered important to patients and clinicians according to the clinical experts and patient and clinician group inputs. In addition, PFS was used to inform the pharmacoeconomic model submitted to CDA-AMC.

Patients were considered to have disease progression if there was an increase of 25% from the lowest confirmed response value in 1 or more of the following based on the IMWG response criteria:⁴²

• Serum M protein absolute increase (must be at least 0.5 g/dL)

• Serum M protein increase of at least 1 g/dL, if the lowest monoclonal component was at least 5 g/dL

• Urine M protein (absolute increase must be at least 200 mg per 24 hours)

• In patients without measurable serum and urine M protein levels, the difference between involved and uninvolved FLC levels (absolute increase must be greater than 10 mg/dL)

• In patients without measurable serum and urine M protein levels and without measurable involved FLC levels, bone marrow plasma cell percentage irrespective of baseline status (absolute increase must be at least 10%)

• Definite increase in the size of existing bone lesions or soft tissue plasmacytomas

• Definite development of new bone lesions or soft tissue plasmacytomas

• At least a 50% increase in circulating plasma cells (minimum of 200 cells per microlitre) if this is the only measure of disease

• Development of plasma cell leukemia

CR or Better Rate

The CR or better rate was a major secondary end point for the CARTITUDE-4 trial, which was defined as the proportion of patients who experienced a CR or sCR. The CR or better rate was considered important to clinicians according to the clinical experts and clinician group inputs. The definitions of CR and sCR were based on the IMWG response criteria:⁴²

• CR: Negative immunofixation on the serum and urine and disappearance of any soft tissue plasmacytomas and less than 5% plasma cells in bone marrow aspirates. In patients whose serum and urine M protein are not measurable, a normal FLC ratio is required.

• sCR: CR as defined above plus a normal FLC ratio and an absence of clonal cells in bone marrow biopsy by immunohistochemistry (kappa-lambda ratio ≤ 4:1 or ≥ 1:2 for patients with kappa and lambda light chains, respectively, after counting ≥ 100 plasma cells).

VGPR or Better Rate

The VGPR or better rate was defined as the proportion of patients who experienced a VGPR or better (i.e., sCR, CR, or VGPR). Response to treatment was analyzed by a validated computerized algorithm. The VGPR or better rate was considered important to clinical experts. The definition of VGPR was based on the IMWG response criteria:⁴²

• Serum and urine M protein are detectable by immunofixation but not on electrophoresis or there is at least a 90% reduction in serum M protein plus a urine M protein level of less than 100 mg per 24 hours.

• If the serum and urine M protein are not measurable, a reduction of at least 90% in the difference between involved and uninvolved FLC levels is required (i.e., a greater than 90% reduction in serum M protein plus a urine M protein of less than 100 mg per 24 hours).

Overall Response Rate

ORR was a major secondary end point for the CARTITUDE-4 trial and was defined as the proportion of patients who experienced a PR or better (i.e., sCR, CR, VGPR, or PR). Response to treatment was analyzed by a validated computerized algorithm. The definition of PR was based on the IMWG response criteria:⁴²

• At least a 50% reduction of serum M protein plus a reduction in 24-hour urinary M protein by at least 90% or to less than 200 mg per 24 hours

• If the serum and urine M protein are unmeasurable, at least a 50% decrease in the difference between involved and uninvolved FLC levels in place of the M protein criteria

• If the serum and urine M protein are unmeasurable and the serum free light assay is also unmeasurable, at least a 50% reduction in plasma cells in place of M protein, provided the baseline bone marrow plasma cell percentage was at least 30%.

In addition to these criteria, if soft tissue plasmacytomas were present at baseline, at least a 50% reduction in their size (sum of the products of the maximal perpendicular diameters of measured lesions) was also required.

Overall MRD Negativity Rate

Overall MRD negativity rate was a major secondary end point for the CARTITUDE-4 trial and was defined as the proportion of patients who have MRD-negative status (at 10^-5) by bone marrow aspirate after the date of randomization and before the start of subsequent antimyeloma therapy. Achievement of MRD negativity has been associated with depth of clinical response and prolongation of PFS and OS.⁴³ MRD status results were reported based on next-generation sequencing and postrandomization assessment. The overall MRD negativity rate was considered important to clinical experts.

Overall Survival

OS was assessed as a major secondary end point in the CARTITUDE-4 trial and was defined as the time from the date of randomization to the date of the patient’s death due to any cause. Patients who were lost to follow-up were censored at the time of loss to follow-up. Patients who died after consent withdrawal were considered as having an OS event. If the patient was alive at the cut-off date for the analysis or their survival status was unknown, then the patient’s data were censored at the date the patient was last known to be alive. The date they were last known to be alive was determined by the maximum collection or assessment date from among selected data domains within the clinical database. OS was considered important to clinicians according to the clinical experts and clinician group input. In addition, OS was used to inform the pharmacoeconomic model submitted to the drug agency.

Duration of Response

DOR was assessed among patients who experienced a PR or better from the date of initial documentation of a response of sCR, CR, VGPR, or PR until the date of first documented evidence of progressive disease based on the computerized algorithm, according to the IMWG response criteria, or until death due to any cause, whichever occurred first. Patients who had not experienced disease progression and were alive were censored at the last disease evaluation before the start of subsequent antimyeloma therapy. DOR was considered important to clinical experts.

Health-Related Quality of Life

Time to Worsening of Symptoms in the MySIm-Q Total Symptom Score

HRQoL, assessed using time to worsening of symptoms according to the MySIm-Q total symptom score, was a major secondary end point in the CARTITUDE-4 trial and was defined as the interval from the date of randomization to the start date of worsening in the MySIm-Q total symptom score. Worsening was defined as a worsening in the given minimal important difference threshold compared to baseline without subsequent improvement to a score above this level. The minimal important difference proposed by the sponsor was a decrease in score that is at least half of the standard deviation from the baseline values, where standard deviation was calculated from the scores at baseline combining both treatment groups. Death due to disease progression was considered as worsening. Patients whose data did not meet the definition of worsening (e.g., a patient experienced worsening by at least the minimal important difference but experienced an improvement and showed no worsening afterward) were censored at the last assessment date of the MySIm-Q. HRQoL was considered important to patients and clinicians according to the clinical experts and patient and clinician group inputs. Refer to Table 7 for details of the MySIm-Q. However, the MySIm-Q instrument is a newly developed questionnaire and specific for MM; no other assessment (e.g., quantitative measures) is available. Furthermore, there is only 1 statistical method, lacking distribution-based methods or anchor-based methods to confirm the MIDs.

Safety Outcomes

In the CARTITUDE-4 trial, adverse events were coded using the Medical Dictionary for Regulatory Activities and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (Version 5.0), with the exception of CRS, which was evaluated according to the American Society for Transplantation and Cellular Therapy grading system. TEAEs were defined as adverse events with onset during the treatment phase or that were a consequence of a pre-existing condition that had worsened since baseline. Notable harms included CRS, neurotoxicity (including ICANS), B-cell aplasia, hypogammaglobulinemia, and immune suppression.

Table 7: Summary of Patient-Reported Outcome Measures and Their Measurement Properties

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; MID = minimal important difference; MM = multiple myeloma; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire.

Statistical Analysis

Analysis of the clinical end points in the CARTITUDE-4 trial is summarized in Table 8.

Sample Size and Power Calculation

The sample size calculation was based on the assumption that cilta-cel can reduce the risk of progressive disease or death by 35% (i.e., an HR of 0.65 for cilta-cel versus SOC, which translated to a median PFS of 20 months for the cilta-cel group, assuming the median PFS for the SOC group was 13 months). Approximately 400 participants (200 per treatment group) were randomized. A total of 250 PFS events was required to achieve approximately 90% power to detect an HR of 0.65 with a log-rank test (2-sided alpha of 0.05). The sample size calculation took into consideration an estimated annual dropout rate of 5% and 1 interim analysis for PFS, to be performed once approximately 188 PFS events (i.e., 75% of the total planned PFS events) had been observed.

Statistical Test or Model

Analysis of the primary end point, PFS, was based on the intention-to-treat (ITT) analysis set, and the Kaplan-Meier method was used to estimate the distribution of overall PFS for each treatment group. Both treatment groups received the same bridging therapy for approximately 2 cycles (about 8 weeks) after randomization; no separation of the Kaplan-Meier curves was expected a priori. Given that a regular log-rank test would lead to a loss of power in such a scenario of nonproportional hazards, the prespecified primary analysis of the PFS end point used a constant piecewise weighted (CPW) log-rank test.

Prespecified primary analysis using the CPW log-rank test was performed by applying a weight of 0 for the first 8 weeks postrandomization and a weight of 1 afterward in the log-rank statistic. The P value from a stratified CPW log-rank test was reported. The HR for the cilta-cel group versus the SOC group and its 95% CI were estimated based on a stratified Cox regression model, with treatment as the sole explanatory variable. Stratification factors used in the stratified analyses included physician’s choice of PVd or DPd, ISS disease stage (I, II, or III), and number of prior lines of therapy (1 versus 2 or 3).

Multiple Testing Procedure

The null hypothesis of no difference between the 2 treatment groups was tested at the 0.05 significance level (overall). The exact significance level for superiority in the PFS interim analysis was determined by the observed number of events using the O’Brien-Fleming boundaries as implemented by the Lan-DeMets alpha spending method. Assuming 188 PFS events were observed at the interim analysis, the 2-sided alpha was to be spent in the interim analysis is 0.0193, and it would be 0.0442 for the final analysis. If the observed 2-sided P value was smaller than 0.0193 or 0.0442 at the interim or final analysis, respectively, the superiority of cilta-cel over PVd or DPd with respect to PFS would be established.

A hierarchical procedure was used to account for multiplicity incurred from testing the primary and secondary end points. Each hypothesis for secondary outcomes was only tested if the null hypotheses for the primary outcome and all preceding secondary outcomes were rejected. The secondary efficacy end points were tested in a hierarchal manner, with a 2-sided significance level of 0.05 (overall), in the following order:

1. CR or better rate

2. ORR

3. overall MRD negativity rate

4. OS

5. time to worsening of symptoms in the MySIm-Q total symptom score.

Data Imputation Methods

No data imputation methods were reported for the CARTITUDE-4 trial.

Subgroup Analyses

Subgroup analyses were performed on the primary efficacy end point of PFS in the ITT analysis set. Subgroups analyses did not take multiplicity into account. The following subgroups are considered relevant based on input from the clinical experts consulted for this review:

• patients with 1 prior line of therapy

• patients with ISS disease stage III

• patients with high-risk cytogenetics

• patients’ prior drug exposure

• patients with presence of extramedullary plasmacytomas

• patients’ baseline ECOG PS

• refractory status of patients’ disease to class of drug.

Additionally, analyses of prespecified subgroups based on age, sex, race, geography, prior drug exposure, and additional clinical characteristics (e.g., type of myeloma, tumour burden, bone marrow plasma cell percentage, total CAR-positive viable T cells infused [ × 10⁶ cells], baseline tumour B-cell maturation antigen expression, baseline renal and hepatic function) were conducted.

Sensitivity Analyses

In the CARTITUDE-4 trial, sensitivity and supplementary analyses were performed for PFS as follows:

Sensitivity analyses:

• Standard “unweighted” stratified log-rank test

• Progressive disease based on investigator assessment according to IMWG, with standard “unweighted” stratified log-rank test

• Progressive disease based on investigator assessment according to IMWG, with stratified CPW log-rank test

In addition to the prespecified sensitivity analyses based on investigator assessment, the sponsor conducted an analysis of PFS based on Independent Review Committee assessment with stratified CPW log-rank test as requested by the FDA.

Supplementary analyses:

• Not censored for start of subsequent antimyeloma therapies: The PFS was derived from the algorithm without censoring data due to the start of subsequent antimyeloma therapies for patients who have not experienced progressive disease. This analysis was performed in a similar manner to the primary analysis, and the PFS definition was similar to that used in the primary analysis, except that progression or death that occurred after the start of subsequent antimyeloma therapies for MM was not censored at the last disease assessment before the start of subsequent therapies. If there was no confirmed progressive disease, the patient was censored at the last disease assessment before loss to follow-up or withdrawal of consent to the study.

• Restricted mean survival time: The proportional hazards assumption was checked via graphical method as well as via formal statistical testing (i.e., Grambsch-Therneau test). Only if the test of the proportional hazards assumption failed at a 2-sided significance level of 0.2, was the restricted mean survival time method performed.⁴⁵ The restricted mean survival time was measured by the area under the Kaplan-Meier PFS curve up to the selected time point, which was calculated for each treatment group, where the selected time point was the smaller value of the longest PFS event time observed from either the cilta-cel group or the SOC group. The difference in restricted mean survival time and 95% CI was reported.

• Censored for death due to COVID-19: The PFS definition used in this analysis is similar to that used in the primary analysis, except as relates to censoring data on death due to COVID-19. Death due to COVID-19 was considered as a PFS event. Instead, patients who died due to COVID-19 were censored at the last disease assessment before the death.

Secondary and Other Outcomes

A summary of statistical analysis of efficacy end points is provided in Table 8.

The CR or better rate, VGPR or better rate, ORR, MRD negativity rate, and time to worsening of symptoms in the MySIm-Q total symptom score were calculated for each treatment group in the ITT analysis set. For each outcome, the corresponding 95% Clopper-Pearson exact CI was provided. The stratified Cochran-Mantel-Haenszel estimate of the OR and its 95% CI and P value was used to test if the CR or better rate, VGPR or better rate, ORR, or MRD negativity rate was the same between the 2 treatment groups. The stratification factors used in these analyses (except for the analysis of the VGPR or better rate) included physician’s choice of PVd or DPd, ISS disease stage (I, II, or III), and number of prior lines of therapy (1 versus 2 or 3). Of note, the analysis of the VGPR or better rate was not included in the hierarchical testing procedure; thus, the results were considered supportive.

Four analyses for OS were planned. The first and second interim analyses for OS were to be performed at the same time as the interim and final analysis for PFS. The third interim analysis for OS was to be performed when approximately 200 OS events had occurred. The final OS analysis was to be performed when approximately 250 OS events had occurred, which also marks the end of study. If the median OS for the SOC group is 31 months and being in the cilta-cel group reduces the risk of death by 30% (HR = 0.7), this translates into a median OS of 44.3 months for the cilta-cel group. The alpha spending function for OS used the power (Kim-DeMets) spending function with a parameter of 2. For example, if the observed number of OS events is 114 at the time of the first interim analysis for PFS (i.e., approximately 188 PFS events), the alpha to be spent for OS is 0.0104 (2 sided). This report focuses on the OS results from the interim analysis (data cut-off date: November 1, 2022), as results from other analyses were not available at the time of this review. The analysis consisted of a stratified log-rank test for the comparison of the OS distribution between the 2 treatment groups, stratified by 3 factors: physician’s choice of PVd versus DPd, ISS disease stage at screening (I, II, or III), and number of prior lines of therapy (1 versus 2 or 3). The Kaplan-Meier method was used to estimate the distribution of overall OS for each treatment. The treatment effect (i.e., HR) and its 2-sided 95% CI were estimated using a stratified Cox regression model with treatment as the sole explanatory variable.

The analysis of DOR was based on patients in the ITT analysis set who experienced a response (PR or better). The distribution of DOR was estimated using the Kaplan-Meier method. Given that DOR was calculated for a subset of patents who had a PR or better, the ITT principle was not strictly followed, and no formal statistical comparison of DOR between the treatment groups was made.

The analysis of time to worsening of symptoms in the MySIm-Q total symptom score was performed in a similar manner as described for PFS.

Analysis Populations

A summary of the analysis populations used in the CARTITUDE-4 trial that are relevant to this review is provided in Table 9.

Table 8: Statistical Analysis of Efficacy End Points

CPW = constant piecewise weighted; CR = complete response; DPd = daratumumab-pomalidomide-dexamethasone; IMWG = International Myeloma Working Group; ISS = International Staging System; MRD = minimal residual disease; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire; NA = not applicable; PVd = pomalidomide-bortezomib-dexamethasone; sCR = stringent complete response; VGPR = very good partial response; vs. = versus.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Table 9: Analysis Populations of the CARTITUDE-4 Trial

ITT = intention to treat.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Results

Patient Disposition

A total of 516 patients were screened of which 97 (23.2%) were excluded at the screening phase. A total of 419 patients were randomized: 208 patients in the cilta-cel group and 211 patients in the SOC group. Of the 419 patients, 416 received the study treatment and comprised the safety analysis set (208 patients in the cilta-cel group and 208 patients in the SOC group); 3 patients were randomized to the SOC group but did not receive treatment. A lower percentage of patients in the cilta-cel group discontinued the study treatment than in the SOC group (15.4% versus 63.0% for cilta-cel versus SOC). The most common primary reasons for treatment discontinuation were progressive disease (14.4% versus 56.3% for cilta-cel versus SOC) and death (1.0% versus 2.4%). Of the 32 patients in the cilta-cel group who discontinued the study treatment, 30 patients (14.4% of the overall cilta-cel group) discontinued on or after bridging therapy and before the start of the conditioning regimen and 2 (1.0%) discontinued on or after the conditioning regimen and before cilta-cel infusion. The most commonly reason for treatment discontinuation among these 32 patients was progressive disease (30 patients [14.4% of the overall cilta-cel group]). As of the data cut-off date of November 1, 2022, a higher percentage of patients were still on the study treatment in the cilta-cel group than in the SOC group (81.3% versus 75.8%).

Table 10: Summary of Patient Disposition From CARTITUDE-4 Trial (All-Consented Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; ITT = intention to treat; SOC = standard of care.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Major Protocol Deviations

In the CARTITUDE-4 trial at the interim analysis (data cut-off date: November 1, 2022), the incidence of important protocol deviations was slightly higher in the cilta-cel group than in the SOC group (█████ ███ ████ for cilta-cel versus SOC) (Table 11). Overall, commonly reported important protocol deviations were received wrong treatment or incorrect dose █████ ███ █████ and other █████ ███ ██████ which were reported more frequently in the cilta-cel group than in the SOC group.

Table 11: Summary of Important Protocol Deviations From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; ITT = intention to treat; SOC = standard of care.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

Baseline Characteristics

At baseline, overall, the demographic characteristics were well balanced between the 2 treatment groups. The median age of all study participants was 61.0 years with a range of 27 to 80 years. Most of the study participants (57.3%) were male, while 42.7% were female. Most participants were white (74.9%), 8.6% of participants were Asian, 3.1% of participants were Black, 0.5% of participants were of another racial group, and 12.9% of participants did not report their racial group information. Participants were enrolled from 81 sites across 16 countries in Europe (61.3%), North America (15.3%), and other regions (23.4%). At baseline, most of the study participants (56.1%) had an ECOG PS of 0.

Overall, the baseline disease characteristics were balanced between the 2 treatment groups. At baseline, most of the participants were at ISS disease stage I (64.0%), had received 2 lines of therapy (40.9%), had at least 1 high-risk cytogenetic abnormality (61.2%), with gain/amp(1q) being the most commonly reported abnormality (47.0%) among all patients. Generally, the baseline disease history was balanced across the treatment groups.

The baseline characteristics outlined in Table 12 are limited to those that are most relevant to this review or were considered likely to affect the outcomes or interpretation of the study results by the review team.

Table 12: Summary of Baseline Characteristics From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; ECOG PS = Eastern Cooperative Oncology Group performance status; FISH = fluorescence in situ hybridization; Ig = immunoglobulin; ISS = International Staging System; ITT = intention to treat; MM = multiple myeloma; SD = standard deviation; SOC = standard of care.

Notes: The ITT analysis set consists of the patients who were randomized in the study. Baseline measurement is defined as the last nonmissing measurement before the initiation of study treatment.

^aThe latest nonmissing ECOG PS score on or before apheresis or cycle 1 day 1 is used. All patients met the inclusion criteria of having an ECOG PS score of 0 or 1 before randomization.

^bISS disease stage is derived based on serum beta-2 microglobulin and albumin.

^cMaximum value from bone marrow biopsy and bone marrow aspirate is selected if both results are available.

^dCytogenetic risk abnormalities are based on central FISH testing, or local FISH and karyotype testing if central FISH testing not available.

Sources: 2023 primary Clinical Study Report for CARTITUDE-4 trial;¹⁴ sponsor-provided additional data.¹⁵

Exposure to Study Treatments

In the CARTITUDE-4 trial, at the interim analysis (data cut-off date: November 1, 2022), all 208 patients (100.0%) randomized to the cilta-cel group had received bridging therapy (PVd or DPd) (Table 13). Of those, 176 patients ███████ had received the conditioning regimen of cyclophosphamide and fludarabine infusion followed by cilta-cel infusion as study treatment. The median duration of study treatment for the SOC group was ███ ██████ ███████ ███ ██ ████ ██████ for the 26 patients who received PVd and ████ ██████ ███████ ███ ██ ████ ██████ for the 182 patients who received DPd. The median total relative dose intensity for individual drugs was as follows: bortezomib █████ ███████ █████ ██ ███████ in the cilta-cel group and █████ ███████ █████ ██ ███████ in the SOC group; pomalidomide: █████ ███████ █████ ██ ███████ in the cilta-cel group and █████ ███████ █████ ██ ██████ in the SOC group; daratumumab: █████ ███████ █████ ██ ███████ in the cilta-cel group and █████ ███████ █████ ██ ███████ in the SOC group; dexamethasone: █████ ███████ █████ ██ ███████ ██ the cilta-cel group and █████ ███████ █████ ██ ███████ in the SOC group.

Table 13: Summary of Patient Exposure From CARTITUDE-4 Trial (Safety Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; DPd = daratumumab-pomalidomide-dexamethasone; NA = not applicable; NR = not reported; PVd = pomalidomide-bortezomib-dexamethasone; SD = standard deviation; SOC = standard of care.

Sources: 2023 primary Clinical Study Report for CARTITUDE-4 trial;¹⁴ sponsor-provided additional data.¹⁵

Prior Therapies

A summary of the types of prior therapies used in the ITT analysis set is provided in Table 14. ███ ██ ███ ███████ patients in the cilta-cel group and ███ ██ ███ ███████ patients in the SOC group had received 1 or more prior transplant. Generally, the most frequently reported prior systemic therapies (≥ 20% of patients in either treatment group) at baseline were similar between the treatment groups, except for melphalan ██████ ███ █████, which was reported more frequently in the SOC group, and thalidomide █████ █ ███ ████ ███, which was reported more frequently in the cilta-cel group.

Table 14: Summary of Prior Therapies From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; ITT = intention to treat; SOC = standard of care.

Sources: 2023 primary Clinical Study Report for CARTITUDE-4 trial;¹⁴ sponsor-provided additional data.¹⁵

Refractory Status to Prior Therapies

The refractory status of patients’ disease to prior therapies by treatment group in the CARTITUDE-4 trial in the ITT set (data cut-off: November 1, 2022) is presented in Table 15. The refractory status to prior therapies was similar in the cilta-cel and SOC groups. All patients (100%) in the cilta-cel and SOC groups had disease that was refractory to a prior line of therapy at some point. The most reported therapies were PI (49.5% versus 45.5% for cilta-cel versus SOC), any PI plus an IMiD (49.5% versus 45.5%), and any anti-CD38 mAB (24.0% versus 21.8%). Most patients (98.6% in both groups) had disease that was refractory to their last line of prior therapy; all patients in both groups had disease that was refractory to lenalidomide.

Table 15: Summary of Refractory Status to Prior Multiple Myeloma Therapy From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; IMiD = immunomodulatory drug; ITT = intention to treat; mAB = monoclonal antibody; PI = proteasome inhibitor; SOC = standard of care.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

Concomitant Medications

A summary of the concomitant medications used in the ITT analysis set is provided in Table 16. All patients (100%) in the cilta-cel and SOC groups received at least 1 concomitant medication. At baseline, generally, the most frequently reported concomitant medications (≥ 30% of patients in either treatment group) were used by a higher proportion of patients in the cilta-cel group than the SOC group, except for nucleosides and nucleotides excluding reverse transcriptase inhibitors ██████ ███ █████ for cilta-cel versus SOC, respectively), proton pump inhibitors ██████ ███ ███████ other viral vaccines ██████ ███ ███████ and potassium █████ ███ ███████, which were reported similarly between groups, and bisphosphonates ██████ ███ ██████ and osmotically acting laxatives██████ ███ ███████, which were reported less frequently in the cilta-cel group.

Table 16: Summary of Concomitant Medications From CARTITUDE-4 Trial (Safety Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; SOC = standard of care.

Source: Sponsor-provided additional data.¹⁵

Subsequent Treatment

In the interim analysis of the ITT set (data cut-off date: November 1, 2022), fewer patients in the cilta-cel group received subsequent antimyeloma therapy than in the SOC group (█████ ███ ██████ (Table 17). With respect to the most reported subsequent treatment for MM (≥ 5% of patients in either group), higher proportions of patients were observed in the SOC group, except for cilta-cel (████ ███ |). This is because for patients in the cilta-cel group who experienced disease progression before receiving cilta-cel and went on to receive cilta-cel, the cilta-cel infusion was considered as part of their subsequent therapy.

Table 17: Summary of Subsequent Treatment From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

CAR = chimeric antigen receptor; cilta-cel = ciltacabtagene autoleucel; ITT = intention to treat; SOC = standard of care.

^aFor patients in the cilta-cel group who experienced disease progression before receiving cilta-cel and went on to receive cilta-cel, the cilta-cel infusion was considered as part of their subsequent therapy.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴ Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Efficacy

Progression-Free Survival

In the interim analysis (data cut-off date: November 1, 2022), the median duration of follow-up for PFS was 15.8 months (range, 0.2 months to 27.3 months) in the cilta-cel group and 15.3 months (range, 0 months to 25.4 months) in the SOC group (Table 18). Patients in the cilta-cel group had fewer PFS events, as assessed using a computerized algorithm, than in the SOC group. Sixty-five patients (31.3%) in the cilta-cel group and 122 patients (57.8%) in the SOC group experienced an event; among them, 48 (23.1%) experienced disease progression and 17 (8.2%) died in the cilta-cel group, and 118 (55.9%) experienced disease progression and 4 (1.9%) died in the SOC group. The median PFS was not reached (95% CI, 22.8 months to not estimable) for the cilta-cel group and was 11.8 months (95% CI, 9.7 months to 13.8 months) for the SOC group. The Kaplan-Meier estimate of PFS probability at 12 months was 75.9% (95% CI, 69.4% to 81.1%) for the cilta-cel group and 48.6% (95% CI, 41.5% to 55.3%) for the SOC group; the between-group difference was █████ ████ ███ █████ ██ ██████. At 24 months, the Kaplan-Meier estimate of PFS probability was █████ ████ ███ █████ ██ ██████ for the cilta-cel group and █████ ████ ███ ████ ██ ██████ for the SOC group; the between-group difference was not reported. The interim analysis showed an improvement in PFS for patients receiving cilta-cel compared with SOC (HR = 0.26; 95% CI, 0.18 to 0.38; P < 0.0001). This means that patients in the cilta-cel group were 74% less likely to experience of death or progression than patients in the SOC group. The Kaplan-Meier estimate of the PFS distribution among the interim analysis population is depicted in Figure 2.

The results for all planned sensitivity analysis — including PFS based on computerized algorithm analyzed by the standard unweighted stratified log-rank test (HR = 0.40; 95% CI, 0.29 to 0.55; nominal P < 0.0001), PFS based on investigator assessment of disease progression (HR = 0.39; 95% CI, 0.28 to 0.52; nominal P < 0.0001), standard unweighted stratified CPW (HR = 0.25; 95% CI, 0.18 to 0.37; nominal P < 0.0001), and PFS based on Independent Review Committee assessment of disease progression (HR = 0.26; 95% CI, 0.18 to 0.38; nominal P < 0.0001) — were consistent with the primary analysis across all prespecified and additional sensitivity analyses.⁴⁶ Subgroup analyses of PFS based on a computerized algorithm in the primary analysis were consistent with the primary analysis across all prespecified subgroups. Refer to Appendix 1 for the detailed subgroup analyses data.

Figure 2: Kaplan-Meier Estimates of Progression-Free Survival in the Interim Analysis (ITT Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; DPd = daratumumab-pomalidomide-dexamethasone; ITT = intention to treat; PVd = pomalidomide-bortezomib-dexamethasone.

Note: Arm A = PVd or DPd; Arm B = A sequence of apheresis, bridging therapy (PVd or DPd), conditioning regimen (cyclophosphamide and fludarabine), and cilta-cel infusion.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

CR or Better Rate

In the interim analysis (data cut-off date: November 1, 2022), the CR or better (sCR or CR) rate determined by computerized algorithm was higher in the cilta-cel group than in the SOC group (73.1% versus 21.8% for cilta-cel versus SOC) (Table 18). Among patients who experienced CR or better, 58.2% (95% CI, 51.2% to 65.0%) of patients in the cilta-cel group and 15.2% (95% CI, 10.6% to 20.7%) of patients in the SOC group had an sCR, and 14.9% (95% CI, 10.4% to 20.5%) of patients in the cilta-cel group and 6.6% (95% CI, 3.7% to 10.9%) of patients in the SOC group had a CR. The stratified Cochran–Mantel–Haenszel estimate of the OR was 10.3 (95% CI, 6.5 to 16.4; P < 0.0001). Similar results for CR or better rate were observed for the sensitivity analysis in which determination of response was based on investigator assessment.

VGPR or Better Rate

In the interim analysis (data cut-off date: November 1, 2022), ███ ███████ patients in the cilta-cel group and ██ ███████ patients in the SOC group reported a VGPR or better (i.e., sCR, CR, or VGPR). The stratified Cochran–Mantel–Haenszel estimate of OR was ███ (95% CI, ███ ██ ███; nominal P < 0.0001). Among patients who experienced a VGPR or better, 8.2% (95% CI, 4.8% to 12.8%) of patients in the cilta-cel group and 23.7% (95% CI, 18.1% to 30.0%) of patients in the SOC had a VGPR. Similar results for VGPR or better rate were observed for the sensitivity analysis in which determination of response was based on investigator assessment.

Overall Response Rate

In the interim analysis (data cut-off date: November 1, 2022), the ORR (sCR, CR, VGPR, or PR) determined by computerized algorithm was higher in the cilta-cel group than in the SOC group (84.6% versus 67.3% for cilta-cel versus SOC). The stratified Cochran–Mantel–Haenszel estimate of the OR was 3.0 (95% CI, 1.8 to 5.0; P < 0.0001). The sections on the CR or better rate and the VGPR or better rate outline how many patients experienced an sCR, CR, or VGPR; 3.4% (95% CI, 1.4% to 6.8%) of patients in the cilta-cel group and 21.8% (95% CI, 16.4% to 28.0%) of patients in the SOC group experienced a PR. Similar rates of overall response were observed for the sensitivity analysis in which determination of response was based on investigator assessment.

Overall MRD Negativity Rate

A higher proportion of patients were reported to have negative overall MRD in bone marrow determined by next-generation sequencing in the cilta-cel group than in the SOC group (60.6% versus 15.6% for cilta-cel versus SOC; OR = 8.7; 95% CI, 5.4 to 13.9; P < 0.0001) in the interim analysis (data cut-off: November 1, 2022).

Overall Survival

At the time of the interim analysis (data cut-off: November 1, 2022), the median OS had not been reached in the cilta-cel group and was 26.7 months (95% CI, 22.5 months to not estimable) in the SOC group. With a median follow-up of 16.0 months for the cilta-cel group and 15.9 months for the SOC group, there were 39 deaths observed in the cilta-cel group (18.8% of patients) and 47 deaths observed in the SOC group (22.3% of patients). The HR was 0.78 (95% CI, 0.50 to 1.20; P = 0.2551). The Kaplan-Meier estimate of OS probabilities decreased from 84.1% (95% CI, 78.4% to 88.4%) to █████ ████ ███ █████ ██ ██████ in the cilta-cel group and from 83.6% (95% CI, 77.8% to 88.0%) to █████ ████ ███ █████ ██ ██████ in the SOC group from 12 to 24 months. A supplementary OS analysis that censored patients who died due to COVID-19 showed results consistent with the primary OS analysis, with an HR of 0.64 (95% CI, 0.41 to 1.02; nominal P = 0.0598), favouring the cilta-cel group. The Kaplan-Meier estimate of the OS distribution among the interim analysis population is depicted in Figure 3.

While the median OS had not been reached at the time of the interim analysis (data cut-off: November 1, 2022), the sponsor provided a descriptive update of OS for the ITT analysis set (based on the survival sweep dated December 13, 2023, which was performed at the request of the European Medicines Agency), with a median follow-up of 28.7 months. The update indicated a trend favouring OS benefit for the cilta-cel group compared to the SOC group, with a HR of 0.57 (95% CI, 0.40 to 0.83). Although the median OS was not reached at this time, the estimated 24-month OS probabilities were 78.8% (95% CI, 72.6% to 83.8%) for the cilta-cel group and 66.2% (95% CI, 59.3% to 72.2%) for the SOC group.⁴⁷ Please refer to Appendix 1 for the Kaplan-Meier estimate of the OS distribution based on the survival sweep.

Figure 3: Kaplan-Meier Plot for Overall Survival (ITT Analysis Set; Data Cut-Off: November 1, 2022)

ITT = intention to treat.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

Duration of Response

At the time of the interim analysis (data cut-off: November 1, 2022), the median DOR had not been reached in the cilta-cel group and was ████ █████ █████ ██████ ██ ███ ██████████ in the SOC group. The median follow-up was13.7 months for the cilta-cel group and 14.3 months for the SOC group. Of the patients who experienced a PR or better (176 versus 142 for cilta-cel versus SOC), 143 patients (81.3%) in the cilta-cel group and 80 patients (56.3%) in the SOC group were censored. The Kaplan-Meier estimate of event-free probabilities decreased from 84.7% (95% CI, 78.1% to 89.4%) to █████ ████ ███ █████ ██ ██████ in the cilta-cel group and from 63.0% (95% CI, 54.2% to 70.6%) to █████ ████ ███ █████ ██ ██████ in the SOC group from 12 to 24 months. The Kaplan-Meier estimate of the DOR distribution among the interim analysis population is depicted in Figure 4.

Figure 4: Kaplan-Meier Plot for Duration of Response (Patients With a PR or Better; ITT Analysis Set; Data Cut-Off: November 1, 2022)

ITT = intention to treat; PR = partial response.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

Time to Worsening of Symptoms in MySIm-Q Total Symptom Score

At the time of the interim analysis (data cut-off: November 1, 2022), most participants (85.6% versus 78.2% for cilta-cel versus SOC) had been censored (Table 18). The median time to a sustained worsening of MM symptoms was longer for the cilta-cel group (23.7 months) than for the SOC group (18.9 months), with an HR of 0.42 (95% CI, 0.26 to 0.68; nominal P = 0.0003). The analysis of the time to worsening in MySIm-Q total symptom score was not tested formally at the current interim analysis because it follows OS in the hierarchical testing order and OS was not statistically significant at the time of the clinical cut-off (November 1, 2022). The Kaplan-Meier estimate of event-free probabilities decreased from 84.6% (95% CI, 77.7% to 89.6%) to █████ (95% CI, █████ ██ █████) in the cilta-cel group and from 65.6% (95% CI, 55.2% to 74.2%) to █████ (95% CI, █████ ██ █████) in the SOC group from 12 to 18 months. Kaplan-Meier curves for time to worsening in the MySIm-Q total symptom score are depicted in Figure 5.

Figure 5: Kaplan-Meier Plot for Time to Worsening in MySIm-Q: Total Symptom Subscale (ITT Analysis Set; Data Cut-Off: November 1, 2022) [Redacted]

ITT = intention to treat; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire.

Source: 2023 primary Clinical Study Report for CARTITUDE-4 trial.¹⁴

Table 18: Summary of Key Efficacy Results From CARTITUDE-4 Trial (ITT Analysis Set; Data Cut-Off: November 1, 2022)

CI = confidence interval; cilta-cel = ciltacabtagene autoleucel; CPW = constant piecewise weighted; CR = complete response; DOR = duration of response; DPd = daratumumab-pomalidomide-dexamethasone; HR = hazard ratio; ISS = International Staging System; ITT = intention to treat; MRD = minimal residual disease; MySIm-Q = Multiple Myeloma Symptom and Impact Questionnaire; NE = not estimable; NR = not reported; OR = odds ratio; OS = overall survival; PFS = progression-free survival; PR = partial response; PVd = pomalidomide-bortezomib-dexamethasone; sCR = stringent complete response; SOC = standard of care; VGPR = very good partial response.

^aThe median follow-up duration was 15.9 months at the interim analysis; thus, the 24-month PFS or OS data were immature.

^bHR and 95% CI were from a Cox proportional hazards model with treatment as the sole explanatory variable and stratified by physician’s choice (PVd or DPd), ISS disease stage (I, II, III), and number of prior lines (1 vs. 2 or 3) as randomized, including only PFS events that occurred more than 8 weeks postrandomization. A HR less than 1 indicates an advantage for the cilta-cel group.

^cP value was based on the CPW log-rank test (weight = 0 in the log-rank statistic for the first 8 weeks postrandomization, and 1 afterward), stratified by physician’s choice (PVd or DPd), ISS disease stage (I, II, III), and number of prior lines (1 vs. 2 or 3) as randomized.

^dA Cochran-Mantel-Haenszel estimate of the common OR for stratified tables is used. An OR greater than 1 indicates an advantage for the cilta-cel group. The stratification factors are physician’s choice (PVd or DPd), ISS disease stage (I, II, III), and number of prior lines of therapy (1 vs. 2 or 3) as randomized.

^eP value from the Cochran-Mantel-Haenszel chi-square test.

^fExact 95% CI.

^gP value from Fisher exact test.

^hP value was based on the log-rank test stratified by physician’s choice (PVd or DPd), ISS disease stage (I, II, III), and number of prior lines (1 vs. 2 or 3) as randomized.

ⁱDOR was calculated among responders (with a partial or better response) from the date of initial documentation of a response, which means the intention-to-treat principle was not strictly followed. With this, the statistical inference between 2 treatment arms with regarding to DOR is not recommended.

^jThis end point was not tested formally at the current interim analysis because it followed OS in the hierarchical testing order and OS was not significant at the time of clinical cut-off due to immaturity of the data.

Sources: 2023 primary Clinical Study Report for CARTITUDE-4 trial;¹⁴ sponsor-provided additional data.^15,16 Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Harms

Harms data in the CARTITUDE-4 trial interim analysis (data cut-off: November 1, 2022) are summarized in Table 19.

Treatment-Emergent Adverse Events

In the CARTITUDE-4 trial, all patients from both groups experienced at least 1 TEAE. By system organ class, the most common TEAEs were blood and lymphatic system disorders, including neutropenia (89.9% versus 85.1% for cilta-cel versus SOC), anemia (54.3% versus 26.0%), and thrombocytopenia (54.3% versus 31.3% for cilta-cel versus SOC). Other common TEAEs were gastrointestinal disorders, affecting 154 patients (74.0%) in the cilta-cel group and 116 patients (55.8%) in the SOC group, including conditions such as nausea, diarrhea, constipation, and vomiting. The frequency of immune system disorders was notably higher in the cilta-cel group, where they were experienced by 157 patients (77.5%), than in the SOC group, where they were experienced by 17 patients (8.2%).

Serious Adverse Events

Ninety-two patients (44.2%) in the cilta-cel group and 81 patients (38.9%) in the SOC group reported at least 1 SAE. Infections and infestations (24.0% versus 24.5% for cilta-cel versus SOC), including COVID-19 pneumonia (5.8% versus 4.3%), were the most reported SAEs.

Withdrawals Due to Adverse Events

In the CARTITUDE-4 trial, TEAEs leading to withdrawal of any component of study treatment were reported for | ████████ ██████ in the cilta-cel group and ██ ████████ ███████ in the SOC group.

Mortality

TEAEs were reported as the primary cause of death for ██ ████████ ██████ in cilta-cel group and | ████████ ██████ in SOC group. The primary cause of death due to a TEAE was COVID-19 pneumonia for 7 patients (3.4%) in the cilta-cel group and 1 patient (0.5%) in the SOC group. Although the entirety of the trial overlapped with the COVID-19 pandemic, mitigation strategies were put into place via protocol amendment, and no further COVID-19–related mortality was observed following cilta-cel infusion.

Notable Harms

The sponsor and/or the clinical experts identified notable harms as including CRS, neurotoxicity (including ICANS), B-cell aplasia, hypogammaglobulinemia, and immune suppression. CRS was reported for 76.1% of patients in the cilta-cel group (134 of 176 patients who received cilta-cel as study treatment), with the majority being grade 1 (52.8% of the 176 patients who received cilta-cel as study treatment). Only 2 patients (1.1%) experienced grade 3 CRS, and no grade 4 or 5 CRS was reported. In total, 36 patients (20.5%) from the cilta-cel group experienced CAR T-cell neurotoxicity, including ICANS in 8 patients (4.5% of the 176 patients who received cilta-cel as study treatment). Among the 8 patients with ICANS, 6 patients (3.4% of the 176 patients who received cilta-cel as study treatment) had grade 1 events and 2 patients (1.1%) had grade 2 events. Hypogammaglobulinemia was observed in 88 of 208 patients (42.3%) in the cilta-cel group and 13 of 208 patients (6.3%) in the SOC group, with 15 patients (7.2%) in the cilta-cel group and 1 patient (0.5%) in the SOC group experiencing grade 3 or 4 hypogammaglobulinemia. Immune suppression was observed in ███ ████████ ███████ in the cilta-cel group and ███ ████████ ███████ in the SOC group. No data for B-cell aplasia were reported.

Table 19: Summary of Harms Results From CARTITUDE-4 Trial (Safety Analysis Set; Data Cut-Off: November 1, 2022)

Cilta-cel = ciltacabtagene autoleucel; CRS = cytokine release syndrome; ICANS = immune effector cell–associated neurotoxicity syndrome; NA = not applicable; NR = not reported; SAE = serious adverse event; SOC = standard of care; TEAE = treatment-emergent adverse event.

Notes: Adverse events are reported using Medical Dictionary for Regulatory Activities version 25.0. ICANS was evaluated according to the American Society for Transplantation and Cellular Therapy consensus grading system.

^aTEAEs with at least 10% frequency was reported.

^bThe reason for deaths (not due to disease progression) was reported as an adverse event if the adverse event was treatment emergent; it was reported as “other” if the adverse event was non–treatment emergent.

Sources: 2023 primary Clinical Study Report for CARTITUDE-4 trial;¹⁴ sponsor-provided additional data.^15,16 Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Critical Appraisal

Internal Validity

The CARTITUDE-4 trial is an ongoing phase III, open-label, active-control RCT. The methods of randomization involved stratification by physician’s choice of PVd or DPd, ISS disease stage at screening (I, II, or III), and number of prior lines of therapy (1 versus 2 to 3), which were considered appropriate by the review team. There was generally no notable imbalance in the baseline patient characteristics between treatment groups.

In the ITT analysis set, 32 of the 208 patients (15.4%) in the cilta-cel group did not receive cilta-cel infusion and 3 of the 211 patients (1.4%) in the SOC group did not receive PVd or DPd as their randomly assigned study treatments. Of the 32 patients in the cilta-cel group who did not receive cilta-cel infusion, 30 (14.4%) discontinued study treatment on or after bridging therapy and before the start of the conditioning regimen and 2 (1.0%) discontinued on or after the conditioning regimen and before cilta-cel infusion. The most common reason for discontinuing the study treatment was disease progression (30 patients [14.4%]). Bridging therapy included PVd or DPd, per physician’s choice. The differential imbalance in the baseline characteristics of the patients who discontinued treatment between the 2 groups could be a source of attrition bias against the cilta-cel group. This would complicate the interpretation of causal inference — which may require an appropriate and precise definition of the estimand (i.e., not considering patients who actually did not receive the cilta-cel treatment) — about the use of cilta-cel and improvement in efficacy outcomes. The clinical experts confirmed that the median duration of study treatment for the SOC group (PVd: 4.8 months; DPd: 11.8 months) was reflective of clinical practice but highlighted that for patients whose disease is refractory to lenalidomide, the experts would expect a slightly longer duration of treatment for PVd of approximately 8 months. The review team noted that the relatively short study treatment for patients with PVd (26 patients [12.3%]) may bias the study results in favour of cilta-cel. Consistently and notably, higher proportions of patients in the cilta-cel group than in SOC group received concomitant therapies for the control of various clinical symptoms or disorders associated with the increased incidence of adverse events. These therapies included antimicrobial and antiviral medications, normal human immunoglobulin, serotonin (5ht3) antagonists, paracetamol, enoxaparin, interleukin inhibitors, antiepileptics, and drugs for the treatment of psychological and sleeping disorders. In particular, the clinical experts indicated that the higher use of the above-mentioned concomitant medications was due to the increased incidence of TEAEs related to infections, CRS, and gastrointestinal disorders (e.g., nausea and/or vomiting). In particular, the frequency and/or severity of the adverse events — which might have significantly affected patients’ quality of life, including ability to perform daily functions — could have been reduced due to the use of those concomitant medications. Overall, the increased use of concomitant medications might have had an impact on the reported adverse events and HRQoL outcomes in the cilta-cel group. Fewer patients received subsequent treatment in the cilta-cel group than in the SOC group; this would bias the OS results against the cilta-cel group.

As the CARTITUDE-4 trial is ongoing, results were only available from the interim analysis for this review. At the time of the interim analysis, the median PFS and median OS had not been reached in the cilta-cel group, which casts uncertainty on the assessment of treatment effect in terms of both median survival time and HRs. Although results from the sponsor-conducted subsequent OS analysis (data cut-off: December 13, 2023) indicated a trend favouring OS benefit for the cilta-cel group compared to the SOC group, the median OS was still not yet reached at this time. Moreover, the statistical testing of the subsequent OS analysis was not controlled for the overall type I error; therefore, the results were descriptive and should be considered as supportive data.

A multiple testing procedure was employed to control the overall type I error for the primary end point of PFS and the major secondary end points of CR or better rate, ORR, overall MRD negativity rate, OS, and time to worsening of symptoms in the MySIm-Q total symptom score in the interim analysis. Many of the outcomes used in the CARTITUDE-4 trial (PFS, OS, CR or better rate, VGPR or better rate, ORR, and DOR) were identified as clinically important by patients and/or clinicians. However, VGPR or better rate and DOR were not part of the statistical testing strategy and thus were not adjusted for multiple testing; therefore, the ability to draw conclusions from these results may be limited. The clinical experts confirmed that a 35% reduction in the risk of progressive disease or death and a median PFS of 20 months for the cilta-cel group, used in the sample size calculation, was considered to be reasonable and clinically meaningful.

External Validity

According to the clinical experts consulted for this review, the demographic and disease characteristics of the CARTITUDE-4 trial population were generally reflective of the patient population with RRMM who would be candidates for treatment with cilta-cel. However, the trial excluded a small group of patients (less than 5%) with symptomatic MM who do not have measurable disease; their disease may have bone and/or bone marrow involvement, which would be detectable on imaging such as PET or CT scans or biopsies. Those patients who have confirmed relapsed disease, even if nonsecretory, may still benefit from cilta-cel as per the feedback from the clinical experts. The CARTITUDE-4 trial excluded patients with an ECOG PS of 2; the clinical experts would consider those patients to be eligible for cilta-cel as patients with an ECOG PS of 2 are similar to patients with an ECOG PS of 1 or less in terms of responding to the treatment. Additionally, the clinical experts mentioned that ECOG PS scores may change along a patient’s disease course; therefore, there is no reason to exclude patients with an ECOG PS of 2 from treatment with cilta-cel. Careful consideration of overall health and ability to withstand acute toxicities such as CRS would be important. The clinical experts commented that the proportion of patients with high-risk cytogenetic abnormality (61.2%) in the CARTITUDE-4 trial was higher than what they would expect in clinical practice (about 25% to 35% of patients), which may indicate that the study included a higher-risk patient population. The patient population (median age: 61.0 years) was generally younger than the general population with MM. The proportion of patients with a prior autologous stem cell transplant (82.2% versus 87.7% for cilta-cel versus SOC) was higher than what the clinical experts would expect in clinical practice (about 50% to 60% of patients), which may indicate that the study included a favourable patient population that may not be reflective of patients with RRMM in clinical practice.

In the SOC group, 26 patients (10.4%) used PVd and 182 patients (86.3%) used DPd. Although PVd has been recommended by the drug agency for RRMM in patients who have received at least 1 prior treatment regimen including lenalidomide, PVd is not commonly used or an SOC in Canada.³³ The clinical experts estimated that about 10% to 25% of patients whose disease is refractory to lenalidomide might be on PVd in clinical practice in Canada. Moreover, DPd is not currently funded in Canada, but the clinical experts commented that the treatment effect of DPd is similar to IsaPd, which is used in clinical practice in Canada. Additionally, there was no study site in Canada in the CARTITUDE-4 trial. Overall, the review team noted all these factors may compromise the generalizability of the study results in Canada.

At the time this report was prepared, the duration of follow-up (median: 15.9 months) in the interim analysis (data cut-off: November 1, 2022) was adequate for the assessment of the primary efficacy end point of PFS but inadequate for the assessment of OS, as per feedback from the clinical experts. Patients and/or clinicians indicated that prolonging PFS and OS, delaying disease progression, maintaining HRQoL, and controlling the symptoms of the disease were critical considerations. Overall, it is uncertain the extent to which the observed OS, patient-reported HRQoL, and disease symptom results from the CARTITUDE-4 trial could be generalized to clinical practice in Canada considering the limited representativeness of the study population due to restrictive eligibility criteria and comparators that were not exactly reflective of current clinical practice in Canada.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For the pivotal CARTITUDE-4 trial identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for the outcomes considered most relevant to inform the expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group:^12,13

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

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

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

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

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

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The target of the certainty of evidence assessment was the presence or absence of a clinically important effect based on thresholds for PFS informed by the clinical experts consulted for this review; there is no established minimal important difference and the clinical experts could not provide a threshold of important difference, so the target of the certainty of evidence assessment was the presence or absence of any (non-null) effect for CR or better rate, VGPR or better rate, overall MRD negativity rate, OS, DOR, HRQoL, and SAEs.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for cilta-cel versus SOC.

Long-Term Extension Studies

No long-term extension studies were submitted for this review.

Indirect Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CDA-AMC review team.

Objectives for the Summary of Indirect Evidence

The pivotal CARTITUDE-4 trial provided a head-to-head comparison between cilta-cel and physician’s choice of 2 SOC therapies (PVd or DPd) among patients with RRMM. However, no direct evidence was included in the submission to support comparisons of the efficacy or safety of cilta-cel with other available treatments for the patient population under review in Canada. Hence, an ITC was warranted to address this evidence gap.

Description of Indirect Comparisons

The sponsor submitted 2 ITC analyses to compare cilta-cel to relevant treatment comparators in Canada. One ITC report presented analyses of IPD from the pivotal and 3 additional comparator clinical trials — the CANDOR (Kd), CASTOR (Vd, DVd), and APOLLO (Pd) trials — using IPTW methods. The other report presented an unanchored MAIC, using IPD from the pivotal CARTITUDE-4 trial and summary-level data from 2 comparator trials: the ICARIA-MM (IsaPd) and BOSTON (SVd) trials.

Table 20: Study Selection Criteria and Methods for ITCs Submitted by the Sponsor

AE = adverse event; cilta-cel = ciltacabtagene autoleucel; CR = complete response; HRQoL = health-related quality of life; ITC = indirect treatment comparison; MM = multiple myeloma; MRD = minimal residual disease; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PICOS = population, interventions and comparisons, outcomes, and study design; RCT = randomized controlled trial; RRMM = relapsed or refractory multiple myeloma; VGPR = very good partial response.

^aTrials that had at least 75% of patients with RRMM with exposure to prior lenalidomide or a subgroup reporting based on prior lenalidomide exposure or lenalidomide refractoriness were considered in the feasibility assessment.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

ITC Design

Objectives

The objectives of the sponsor-conducted ITCs were to assess the efficacy and/or safety of cilta-cel compared to other relevant treatments available in Canada.

Study Selection Methods

The sponsor conducted a feasibility assessment to understand the appropriateness of conducting ITCs against all approved and commonly used regimens in the US, UK, Spain, France, Germany, Italy, the Netherlands, and Canada. For the purposes of this CDA-AMC review, only comparators relevant to and available in Canada were presented by the sponsor.

A broad systematic literature review was conducted to identify clinical trials assessing the efficacy, safety, and HRQoL associated with cilta-cel compared to other relevant treatments (Table 20) among adult patients with RRMM who had received prior lenalidomide. The inclusion criteria covered single-arm trials and RCTs, published up to May 2022. The following electronic databases were screened: MEDLINE, Embase, Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials, and PsychInfo. Some details of the literature search, including search dates, search terms, and limitations to searches, were not reported.

Screening for potentially eligible articles was conducted by 2 reviewers. Data were extracted by 1 reviewer and validated by an additional reviewer; a third investigator was consulted if there were any disagreements. Quality assessment of the studies evaluated in the feasibility assessment was conducted using the Cochrane Collaboration’s Risk of Bias tool (version 1). Details regarding the systematic literature review protocol were not available in the sponsor’s submission.

The feasibility assessment included evaluation of the network connectivity as well as evaluation of the alignment between each comparator trial and the CARTITUDE-4 trial in terms of study design features, patients’ characteristics, previous treatments, and outcome definitions. A total of 142 trials were identified through the literature search, 11 of which were informing the potential network for the CARTITUDE-4 trial and were assessed in detail for feasibility of used in the ITC. Only 2 relevant comparators, Pd and IsaPd, were identified for potential inclusion in the network for the CARTITUDE-4 trial. Given that the network contained a limited number of comparators, the majority of which were not relevant to the Canadian clinical context, and that there was substantial heterogeneity across studies, the sponsor determined that a network meta-analysis approach would be inappropriate (Table 22).

IPTW and multivariable regression analyses were conducted for the studies with available IPD, specifically for DVd (CASTOR trial), Vd (CASTOR trial), Kd (CANDOR trial), and Pd (APOLLO trial). Unanchored MAICs were conducted for comparisons to IsaPd (ICARIA-MM trial) and SVd (BOSTON trial) using summary-level data for the comparator trials. A summary of the feasibility assessment and recommended analyses is reported in Table 22.

The feasibility assessment further identified limitations for the IKEMA trial (IsaKd), which precluded the conduct of an ITC of cilta-cel versus IsaKd. These limitations included the following:

• Lack of direct and indirect connectivity between the IKEMA trial (IsaKd and Kd groups) and the CARTITUDE-4 trial (cilta-cel and SOC groups).

• Substantial heterogeneity between the CARTITUDE-4 and IKEMA trials:

  • Lenalidomide refractoriness was an inclusion criterion for the CARTITUDE-4 trial; only 32% of patients in the IKEMA trial were had lenalidomide-refractory disease.

  • Patient descriptive and outcomes data specific to the lenalidomide-refractory disease subgroup were not available from the IKEMA trial.

  • The IKEMA trial excluded patients previously exposed to carfilzomib, those whose disease was refractory to carfilzomib, and those whose disease was refractory to anti-CD38 mAB, which corresponds to, respectively, 37.0%, 24.5%, and 24.0% of patients treated with cilta-cel in the CARTITUDE-4 trial.

  • The IKEMA trial did not report OS data.

Since it was not possible to generate comparative estimates for cilta-cel versus IsaKd, the sponsor submitted a deviation request to CDA-AMC to exclude IsaKd as a comparator in the reference case pharmacoeconomic analysis. This was approved by CDA-AMC on March 12, 2024.

ITC Analysis Methods — IPTW

The ITCs included cilta-cel, DVd, Kd, Pd, and Vd treatments from the CARTITUDE-4, CANDOR, CASTOR, and APOLLO trials. The IPTW methods are summarized in Table 22. Patients from the comparator trials were selected for the analyses if they met key inclusion criteria from the CARTITUDE-4 trial (i.e., have received 1 to 3 prior lines of therapy including a PI and an IMiD, had disease that was refractory to lenalidomide, and had an ECOG PS of 0 or 1). Additionally, patients with prior anti-CD38 mAB exposure in the cilta-cel cohort were excluded to align with the exclusion criteria in the daratumumab clinical trials. The sponsor reported that the relative treatment effect of cilta-cel did not differ in the CARTITUDE-4 trial between patients who had and had not previously had anti-CD38 mAB exposure, based on insignificant findings from the statistical analysis assessing the interaction between treatment outcome and anti-CD38 mAB exposure.

IPTW with average treatment effect on the treated (ATT) weighting was selected as the primary analysis for the PFS, ORR, CR or better, and VGPR or better end points. For the analyses of OS, a multivariable regression approach was deemed more appropriate due to the lower number of events and limited sample size of comparator trials. The outcomes of interest are defined in Table 23.

Prognostic baseline characteristics for adjustment in the analyses were identified and ranked in order of importance a priori, based on input from independent clinical experts.^49,50 Refractory status, cytogenetic risk, ISS or revised ISS disease stage, presence of plasmacytomas or extramedullary disease, and time to disease progression on prior line were identified as the prognostic factors of most relevance. Cytogenetics was ultimately excluded from the analyses due to the high proportion of unreported data in the comparator clinical trials (range, 25% to 60%). Hence, the remaining 4 factors constitute the variables used for adjustment in the base-case analysis.

Two sensitivity analyses were conducted, both of which allowed additional adjustments for several covariates: multivariable regression models, and IPTW with average treatment effect on the control (ATC) weights (Table 23).

To ensure balance between the cilta-cel and comparator cohorts, selected baseline characteristics were adjusted for using either propensity score or regression methods.⁵¹ Logistic regression methods were adopted to estimate propensity scores, with treatment as the dependent variable and baseline covariates as explanatory variables. Weights were derived for each participant, based on the propensity scores and by using weighting formulas for the desired target population. In the case of ATT weights, a score of 1 was assigned to patients in the CARTITUDE-4 trial, while patients in the comparator trials were reweighted based on the probability of receiving treatment.⁵² In the case of ATC weighting, a score of 1 was assigned to patients in the comparator treatment cohorts, while patients in the CARTITUDE-4 trial were reweighted based on the probability of receiving treatment.

Relative efficacy was assessed for both the unadjusted (i.e., cilta-cel versus comparative treatment before IPTW) and the adjusted (i.e., with IPTW) comparisons for all outcomes. For binary outcomes (i.e., ORR, VGPR or better, and CR or better), logistic regression (with weights applied for the adjusted comparison) were adopted to estimate outcomes such as RRs, ORs and corresponding 95% CIs. Cox proportional hazards models were applied to estimate HRs and 95% CIs for PFS (with weights applied for the adjusted comparison) and OS (with factors included as model covariates for the adjusted comparison). The appropriateness of the proportional hazards assumption was assessed based on visual inspection of the log-cumulative hazard and Schoenfeld residuals plots and on the performance of the Grambsch-Therneau test. Variables with less than 25% missing values were imputed using the multiple imputation with chained equation. Specifically, missing variables in the CANDOR trial (i.e., time to disease progression on prior line [4.1% missing for DKd and 4.4% for Kd], years since MM diagnosis [5.1% missing for DKd and 2.2% for Kd], and hemoglobin [1% missing for DKd]) were imputed. Imputation was not required for the CARTITUDE-4, CASTOR, or APOLLO trials.

ITC Analysis Methods — Unanchored MAIC

The methodology of unanchored MAICs, conducted using IPD from the CARTITUDE-4 trial and summary-level data from the comparator trials (IsaPd [ICARIA-MM trial] and SVd [BOSTON trial]), is described in Table 23.⁵³

The feasibility assessment covered comparison of the key aspects of each comparator trial — including inclusion and exclusion criteria, general study designs, outcome definitions, and baseline characteristics — to the CARTITUDE-4 trial. Patients in the cilta-cell group from the CARTITUDE-4 trial who satisfied the eligibility criteria from the comparator trials were included in the analyses (Table 21).

Table 21: Eligibility Criteria Matching

Cilta-cel = ciltacabtagene autoleucel; IsaPd = isatuximab-pomalidomide-dexamethasone; SVd = selinexor-bortezomib-dexamethasone.

^aSubgroup of patients whose disease was refractory to lenalidomide.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

The prognostic factors to be used for adjustments in the analysis were identified and ranked by importance a priori, based on input from independent clinical experts consulted by the sponsor. The following prognostic factors were included in the base case: refractory status, cytogenetic risk, and ISS disease stage. Refractory status was not adjusted in the MAIC with SVd as the BOSTON trial did not report these data.

Sensitivity analyses were performed to include an expanded list of adjusting variables (Table 23).

IPD from the CARTITUDE-4 trial were weighted so that the baseline characteristics of the participants matched the summary-level baseline characteristics from the published comparator trials. Propensity score weighting was implemented to assign weights to patients from the CARTITUDE-4 trial by their inverse odds of being in that group versus the comparator cohort. Separate propensity score models were estimated for each of the comparator trials using the generalized method of moments, including baseline risk factors available in the CARTITUDE-4 trial and the comparator trials.⁵⁴

ESS, corresponding to the size of the unweighted sample that would result in the same precision as the weighted cohort of the patients receiving cilta-cel, was reported (Table 22).

Reconstructed IPD were derived from the reported results for the comparators and by simulating data for PFS and OS from digitally scanned, published Kaplan-Meier curves using a previously published and validated algorithm method.⁵⁵

To estimate the relative benefit for cilta-cel versus comparators on survival outcomes (OS and PFS), weighted IPD for cilta-cel from the CARTITUDE-4 trial and simulated IPD for the comparator trials were analyzed for each pairwise comparison separately using weighted logistic regression and weighted Cox proportional hazards regression.

Table 22: Summary of Feasibility Assessment and Analysis Recommendations

ATT = average treatment effect on the treated; DVd = daratumumab-bortezomib-dexamethasone; ESS = effective sample size; IPTW = inverse probability of treatment weighting; ITC = indirect treatment comparison; IsaKd = isatuximab-carfilzomib-dexamethasone; IsaPd = isatuximab-pomalidomide-dexamethasone; Kd = carfilzomib-dexamethasone; MAIC = matching-adjusted indirect comparison; NA = not available; OS = overall survival; Pd = pomalidomide-dexamethasone; PFS = progression-free survival; SVd = selinexor-bortezomib-dexamethasone; Vd = bortezomib-dexamethasone; vs. = versus.

^aPatients with no prior exposure to anti-CD38 mAB.

^bPatients who received 1 to 3 prior lines of therapy including a proteasome inhibitor and an immunomodulatory drug, whose disease was refractory to lenalidomide, and who had an Eastern Cooperative Oncology Group performance status of less than 2.

^cNo common comparator. Imbalance on lenalidomide-refractoriness (32% with lenalidomide-refractory disease), and patient baseline characteristics and subgroup outcomes specific to the lenalidomide-refractory disease subgroup were not available from the IKEMA trial. The IKEMA trial excluded patients previously exposed to carfilzomib, those whose disease was refractory to carfilzomib, and those whose disease was refractory to anti-CD38 mAB. OS was not reported in the IKEMA trial.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Table 23: ITC Analysis Methods

ATC = average treatment effect on the controls; ATT = average treatment effect on the treated; cilta-cel = ciltacabtagene autoleucel; CR = complete response; DSU = Decision Support Unit; DVd = daratumumab-bortezomib-dexamethasone; ECOG PS = Eastern Cooperative Oncology Group performance status; IMiD = immunomodulatory drug; IMWG = International Myeloma Working Group; IPD = individual patient data; IPTW = inverse probability of treatment weighting; ISS = International Staging System; ITC = indirect treatment comparison; Kd = carfilzomib-dexamethasone; MAIC = matching-adjusted indirect comparison; MM = multiple myeloma; NA = not applicable; NICE = National Institute for Health and Care Excellence; ORR = overall response rate; OS = overall survival; Pd = pomalidomide-dexamethasone; PFS = progression-free survival; PI = proteasome inhibitor; PR = partial response; sCR = stringent complete response; Vd = bortezomib-dexamethasone; VGPR = very good partial response.

^aFor OS, a multivariable regression model was deemed most appropriate and presented as the base case. IPTW-ATT and IPTW-ATC approaches were provided as sensitivity analyses.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Results of ITC

Summary of Included Studies

The sponsor’s assessment of the heterogeneity of the 6 studies included in the ITC analyses (the CARTITUDE-4, CANDOR, APOLLO, CASTOR, BOSTON, and ICARIA-MM trials) is presented in Table 24. All studies were randomized, multicentre, open-label, phase III studies, with sample sizes ranging from 304 to 498 patients. Only the BOSTON study allowed treatment switching, where patients receiving Vd switched to SVd in cases of disease progression.

The feasibility assessment revealed differences in eligibility criteria and in baseline patient and treatment characteristics across the included trials. Even though all studies were conducted among patients who had experienced treatment failure or disease progression on at least 1 prior therapy for MM, only the CARTITUDE-4 trial required patients to have disease that was refractory to lenalidomide. To provide comparable analyses, the sponsor leveraged subgroup analyses on patients whose disease was refractory to lenalidomide from comparator studies (corresponding to approximately 32% to 36% of patients from the CANDOR trial, 80% of patients from the APPOLO trial, 28% of patients from the CASTOR trial, and 92% to 94% of patients from the ICARIA-MM trial; no data were available for the BOSTON trial). The CARTITUDE-4 trial enrolled patients who had had 1 to 3 prior lines of treatment, the ICARIA-MM trial enrolled 34% of patients who had had 4 or more prior lines of treatment, and the APOLLO trial enrolled 14% of patients who had had 4 or more prior lines of treatment. About a third of patients from the CARTITUDE-4 trial had had just 1 prior line of treatment (32.1%), while the APOLLO trial included a very limited proportion of patients who had had just 1 prior line of treatment (12%) and the ICARIA-MM trial did not include any such patients.

Assessment of the patients’ baseline characteristics showed that there were imbalances in patients’ performance status, disease stage, and cytogenetic risk. Further details on patient characteristics are provided in Table 24 and Table 25. The majority of the comparator trials had more patients with advanced stages of disease than the CARTITUDE-4 trial (the CANDOR, APOLLO, and CASTOR trials) as well as more patients with lower cytogenetic risk than the CARTITUDE-4 trial (the CANDOR, APOLLO, CASTOR, BOSTON, and ICARIA-MM trials). The CARTITUDE-4 trial did not include patients with an ECOG PS of 2, while the rest of the studies included varying proportions of individuals with ECOG PS of 2, except for the ICARIA-MM trial, for which such data were not available.

The baseline covariates for populations in the CARTITUDE-4 and daratumumab comparator trials, before and after IPTW adjustments are summarized in Table 26 and Table 27. The CARTITUDE-4 study consisted of 208 patients who were treated with cilta-cel. After the exclusion of 53 patients with prior exposure to an anti-CD38 mAB therapy, the cilta-cel cohort for the analysis included 155 patients. Patients from daratumumab trials were selected, based on the eligibility criteria for the CARTITUDE-4 trial (i.e., received 1 to 3 prior lines of therapy, including a PI and an IMiD; ECOG PS < 2; disease refractory to lenalidomide). The comparator treatment populations consisted of the following cohorts: 44 patients treated with DVd (CASTOR trial), 46 patients treated with Vd (CASTOR trial), 46 patients treated with Kd (CANDOR trial), and 92 patients treated with Pd (APOLLO trial).

Before adjustment, there were numerous pairwise imbalances (standardized mean difference > 0.2) in baseline characteristics between the CARTITUDE-4 and daratumumab trial cohorts.⁵⁶ The daratumumab cohorts had more patients with advanced stages of disease than the cilta-cel cohort (5.2% of patients receiving cilta-cel were at ISS disease stage III versus 22.7% of patients receiving DVd, 17.4% of patients receiving Vd, 17.4% of patients receiving Kd, and 19.6% of patients receiving Pd). Moreover, the DVd, Vd, and Pd cohorts enrolled more patients whose disease was more than double refractory than were found in the CARTITUDE-4 trial population (47.1% in the cilta-cel cohort versus 59.1% in DVd cohort, 52.2% in the Vd cohort, and 51.1% in the Pd cohort). In contrast, the Kd cohort enrolled fewer patients whose disease was more than double refractory (43.5% for Kd) than were found in the cilta-cel cohort (47.1%). Regarding time to disease progression on prior lines of therapy, 85% of patients receiving cilta-cel had experienced disease progression after more than 6 months, compared to 84.1%, 76.1%, 69.6%, and 79.3% of patients receiving DVd, Vd, Kd, and Pd, respectively. The presence of plasmacytomas or extramedullary disease was reported among 18.7% of patients receiving cilta-cel, 22% of patients receiving DVd, 20.4% of patients receiving Vd, 6.5% of patients receiving Kd, and 27.2% of patients receiving Pd. Other differences between the cilta-cel and daratumumab cohorts were observed for number of prior lines of therapy, time since MM diagnosis, hemoglobin levels, prior stem cell transplant, age, prior stem cell transplant, ECOG PS, type of MM, creatinine clearance, sex, and ethnicity.

After adjustment, there were no imbalances, with a standardized mean difference greater than 0.2, in the variables included in the base-case analyses, except for ISS disease stage, for the comparison between the cilta-cel and DVd cohorts (66.5% versus 54.7%, respectively, had ISS stage I disease).

An overview of the baseline patient covariates from the CARTITUDE-4 and comparator trials as well as the CARTITUDE-4 values after adjustment to the comparator studies used in the MAIC analyses are presented in Table 28. The CARTITUDE-4 study initially consisted of 208 patients who were treated with cilta-cel. The numbers of patients in the IsaPd and SVd cohorts were 154 and 53, respectively. Following MAIC adjustment, the ESS of cilta-cel was 26 for the comparison to IsaPd (ICARIA-MM trial) and 188 for the comparison to SVd (BOSTON trial).

When assessing the patient characteristics adjusted in the base-case analysis, there were differences in refractory status of disease (50% of individuals had disease that was refractory to PI in the cilta-cel cohort versus 77% in the IsaPd cohort), cytogenetic risk (59% of the cilta-cel cohort were of high risk versus 19% of the IsaPd and 55% of the SVd cohorts), and ISS disease stage (6% of the cilta-cel cohort were of ISS disease stage III versus 23% of the IsaPd and 8% of the SVd cohorts). Following adjustment, the proportions of individuals across diverse categories of baseline characteristics were balanced across the studies of the MAIC.

Table 24: Assessment of Homogeneity for ITC

DPd = daratumumab-pomalidomide-dexamethasone; ECOG PS = Eastern Cooperative Oncology Group performance status; IMWG = International Myeloma Working Group; IPD = individual patient data; IPTW = inverse probability of treatment weighting; IRC = Independent Review Committee; IsaPd = isatuximab-pomalidomide-dexamethasone; ISS = International Staging System; ITC = indirect treatment comparison; Kd = carfilzomib-dexamethasone; MAIC = matching-adjusted indirect comparison; MM = multiple myeloma; OS = overall survival; Pd = pomalidomide-dexamethasone; PFS = progression-free survival; RRMM = relapsed or refractory multiple myeloma; SOC = standard of care; SVd = selinexor-bortezomib-dexamethasone; Vd = bortezomib-dexamethasone.

Source: Details included in the table are from the sponsor’s summary of clinical evidence.¹⁷

Table 25: Description of Study Inclusion Criteria and Baseline Characteristics From Studies Included in the ITC

1L = first line; 2L = second line; 3L = third line; BOR = bortezomib; cilta-cel = ciltacabtagene autoleucel; DARA = daratumumab; DKd = daratumumab-carfilzomib-dexamethasone; DPd = daratumumab-pomalidomide-dexamethasone; DVd = daratumumab-bortezomib-dexamethasone; ECOG PS = Eastern Cooperative Oncology Group performance status; IMiD = immunomodulatory drug; IsaPd = isatuximab-pomalidomide-dexamethasone; ISS = International Staging System; ITC = indirect treatment comparison; Ixa = ixazomib; K = carfilzomib; Kd = carfilzomib-dexamethasone; LEN = lenalidomide; LOT = line of treatment; NR = not reported; Pd = pomalidomide-dexamethasone; PI = proteasome inhibitor; SOC = standard of care; SVd = selinexor-bortezomib-dexamethasone; Vd = bortezomib and dexamethasone; vs. = versus.

Note: The contents of this table were extracted by the review team from the feasibility assessment documentation for further clarity as a summary of study characteristics was not included in the sponsor’s summary of clinical evidence.

Source: Details included in the table are from the sponsor’s inverse probability of treatment weighting technical report.⁵⁷

Table 26: Overview of Patient Characteristics Before and After IPTW-ATT Weighting (DVd, Vd [CASTOR Trial])

ATT = average treatment effect on the treated; DVd = daratumumab-bortezomib-dexamethasone; ECOG PS = Eastern Cooperative Oncology Group performance status; Ig = immunoglobulin; IPTW = inverse probability of treatment weighting; ISS = International Staging System; MM = multiple myeloma; SMD = standardized mean difference; Vd = bortezomib-dexamethasone.

^aData regarding ethnicity information, including categories reported, were presented as such in the original source (sponsor’s IPTW technical report).

Notes: Shaded cells indicate that these variables were not adjusted for in the given analysis. An SMD between 0 and 0.1 indicates a small difference, an SMD greater than 0.1 and less than or equal to 0.2 indicates a moderate difference, and an SMD of more than 0.2 indicates a substantial difference.⁵⁶ The content of this table was extracted by the review team from the sponsor’s IPTW technical report for further clarity, as details of patient characteristics were not included in the sponsor’s summary of clinical evidence.

Source: Details included in the table are from the sponsor’s IPTW technical report.⁵⁷

Table 27: Overview of Patient Characteristics Before and After IPTW-ATT Weighting (Pd [APPOLO Trial], Kd [CANDOR Trial])