FormalPara Key Summary Points

Why carry out this study?

The clinical prognosis for patients with triple-class exposed relapsed/refractory multiple myeloma (TCE RRMM) has been historically poor, and associated with substantial costs associated with later-line treatment.

Idecabtagene vicleucel (ide-cel, bb2121), a B-cell maturation antigen (BCMA)-directed chimeric antigen receptor (CAR) T cell therapy, showed frequent, deep, and durable responses in patients with TCE RRMM in the pivotal, single-arm, phase 2 KarMMa trial (NCT03361748).

Data regarding the healthcare resource utilization (HCRU) and costs of care in the years following CAR T cell therapy are limited.

What was learned from the study?

Patients treated with a single infusion of ide-cel as part of the phase 2 KarMMa trial used fewer HCRU and had lower costs in the 2 years following treatment.

The estimates of HCRU and costs in the 2 years after treatment with ide-cel suggest that there is a nominal monthly increase in care costs immediately after treatment, and that these costs may be lower in real-world clinical practice settings.

Introduction

Multiple myeloma (MM) is a malignant plasma-cell disorder characterized by a high prevalence of clonal B cells and end-organ damage, such as bone lesions, renal dysfunction, anemia, and hypercalcemia [1, 2]. MM represents approximately 0.9% of new cancer cases and 1.2% of cancer deaths annually according to GLOBOCAN data [3]. In the United States (US), the National Institute of Health’s Surveillance, Epidemiology, and End Results data show a delay-adjusted incidence rate of 7.7/100,000/year [95% confidence interval (CI) 7.6–7.9] for MM, which has risen steadily, with an annual percentage change of 1.1% between 2011 and 2019 [4]. Improved therapeutic options have been reflected in better outcomes over the past decade, and the US mortality rate of 3.0/100,000/year declined between 2014 and 2020 (− 2.0%, 95% CI − 2.7 to − 1.6) [4, 5]. However, the lack of a curative therapy means that most patients who respond to treatment will relapse or become treatment-refractory [5, 6].

Patients with MM are likely to be exposed to multiple treatment modalities during the course of their disease [2, 7]. In patients receiving Medicare, the mean [standard (SD)] total all-cause Medicare payments during subsequent therapy were US$27,557 ($20,300) per patient per month, and the majority of expenditure was due to MM-related treatment cost [8]. The clinical prognosis for patients with relapsed/refractory MM (RRMM) who have received an immunomodulatory agent, a proteasome inhibitor, and anti-CD38 antibody, known to be “triple-class exposed” (TCE), has been historically poor [6, 9,10,11]. For example, a real-world study assessing the clinical outcomes of the anti-CD38 monoclonal antibody daratumumab reported 12-month survival as low as 50% in TCE patients, while, in a real-world study of patients in receipt of Medicare, 40% of TCE patients who initiated a new line of therapy died within 2 years [8, 12]. There is currently no uniform standard of care for TCE patients [8]. An anti-CD38 monoclonal antibody and a proteasome inhibitor was the most common class-level treatment for those TCE patients in receipt of Medicare who added a new line of therapy, while daratumumab in combination with pomalidomide was the most common agent-level regimen [8]. Subsequent lines of treatment intended to reduce recurrence are also associated with fatigue, pain, and a meaningful negative impact on health-related quality of life [13, 14]. From a healthcare resource utilization (HCRU) perspective, substantial costs are associated with these subsequent lines of care [15, 16]. An analysis of insured patients with TCE RRMM in the US who received ≥ 1 post-TCE line of treatment reported mean total all-cause healthcare costs of $722,992 per patient, nearly all of which (91%) were RRMM-related costs ($655,524 per patient) [15].

In recent years, chimeric antigen receptor (CAR) T cell therapies have been approved for use in the US and Europe. Idecabtagene vicleucel (ide-cel, bb2121), a B-cell maturation antigen (BCMA)-directed CAR T cell therapy, is indicated for the treatment of adult patients with RRMM after ≥ 3 and ≥ 4 prior lines of therapy in Europe and the US, respectively [17, 18]. The efficacy and safety of ide-cel in patients with TCE RRMM was assessed in the pivotal, single-arm, phase 2 KarMMa trial (NCT03361748), which was conducted across North America, Europe, and Japan [19, 20]. The primary endpoint was overall response rate; secondary outcomes included further efficacy analyses, safety outcomes, pharmacokinetics, and immunogenicity [20]. Ide-cel showed frequent, deep, and durable responses in patients with TCE RRMM during the KarMMa trial: 73% of all patients had a response (p < 0.001) and 42% had a complete or stringent complete response, with a median duration of response of 10.7 months (95% CI 10.3–11.4) and median progression-free survival of 8.8 months (95% CI 5.6–11.6) [19, 21]. Adverse events were consistent with those observed in the phase 1 clinical study, and included common but transient hematological toxicity and a low incidence of ≥ Grade 3 cytokine-release syndrome (CRS) or neurotoxicity (NT) (6%). Long-term results (median 24.8 months follow-up, data cutoff December 21, 2020) showed a 73% objective response rate and median progression-free survival of 8.9 months in all treated patients [22]. Patients who received the highest target dose (450 × 106 CAR + T cells) had an 81% objective response rate, 39% complete response rate, and a median progression-free survival of 12.2 months [22]. Ide-cel also showed clinically meaningful improvements in several domains of health-related quality of life, including pain and disease symptoms as soon as 1 month after treatment, and for fatigue, cognitive and physical functioning, and global health status by Month 2 [23].

Data regarding the HCRU and costs of care in the years following CAR T cell therapy are limited. Hari et al. [24] analyzed 12-month HCRU and costs for 128 patients in the KarMMa trial, showing a mean post-infusion total healthcare cost of $107,699 per patient that was largely comprised of costs in the first month after treatment (58%). The aim of this retrospective study was to update the previous analysis in order to better capture long-term HCRU by estimating HCRU and total costs of care for patients with RRMM who received an ide-cel infusion in the KarMMa trial over a 24-month post-infusion period.

Methods

The KarMMa Trial

Clinical and patient-reported outcomes from the KarMMa trial have been previously reported [19, 22, 23]. Briefly, patients with TCE RRMM who received ≥ 3 prior antimyeloma regimens and were refractory to their last regimen based on International Myeloma Working Group criteria [25] were enrolled. A total of 128 of 140 (91%) eligible patients received leukapheresis and a single ide-cel infusion (target dose: 150, 300, or 450 × 106 CAR + T cells) [19].

Study Design

HCRU was extracted from the KarMMa clinical trial database and micro-costing, which analyses detailed resource and cost data with the aim of generating precise estimates of economic costs, and is the preferred method for estimating cost of healthcare interventions, was employed [26]. A 2-step methodology was applied to estimate post-infusion cost: (1) to identify resource use, and (2) to apply costs (Fig. 1). Step 1 involved the identification and analysis of HCRU post-ide-cel treatment using the KarMMa clinical trial databases. HCRU categories included: facility care [i.e., outpatient visits, inpatient admissions, inpatient length of stay (LOS): stay (LOS), intensive care unit (ICU) admissions, and ICU LOS], diagnostics (i.e., laboratory tests, imaging), medications [i.e., supportive care, prophylactic treatment, AE management], and procedures (i.e., dialysis, spinal tap, intubation). Step 2 involved the application of unit costs to each HCRU (detailed in Sect. “Determination of HCRU and Costs”). The unit cost was multiplied to account for the frequency and duration of each HCRU. HCRU was assessed from ide-cel administration to a follow-up time of 24 months, excepting resource use related to prophylaxis and AE treatment, which was only assessed in the first 6 months (180 days) post-ide-cel administration in recognition that the focus on AE treatments is primarily for CRS and NT that occur during this time period.

Fig. 1
figure 1

Study design. AE adverse event, ide-cel idecabtagene vicleucel

Full size image

Determination of HCRU and Costs

Post-ide-cel infusion costs were partitioned and calculated by month to account for right-censoring due to loss to follow-up, death, discontinuation, or data cutoff, and then aggregated over a 24-month period. All patients were assumed to have had a minimum 14-day post-infusion inpatient stay according to the KarMMa trial protocol unless the patient transitioned to the ICU during that time. Additional time in the inpatient setting beyond the 14 days was also captured. Outpatient visits were determined by receipt of diagnostics, laboratory tests, or procedures on study days that were not inpatient or ICU days. The total LOS was calculated as the number of inpatient or ICU days. Intubation and extubation were used as a proxy for mechanical ventilation, and lumbar puncture and fluid collection as a proxy for spinal tap. Medication costs related to management of AEs were calculated from the index date through 6 months (Day 180); those for supportive care and prophylactic measures were captured through 24 months (Day 720). Since some patients in the KarMMa trial received ide-cel outside the US, any medication not used in the US was adjusted to the next closest proxy medication used in the US.

Unit costs were applied to each type of HCRU using the US health system (provider) perspective and adjusted to 2022 US dollars using the US Bureau of Labor Statistics Consumer Price Index for medical care (www.bls.gov/cpi/tables/home.htm) and a seasonally adjusted yearly midpoint for estimation. HCRU and costs are provided in Supplementary Material Table S1. Cost per standard inpatient day ($2,944) was estimated from the 2018 Healthcare Cost and Utilization Project (www.hcup-us.ahrq.gov), and cost per ICU day ($9,347) was sourced from Dasta et al. [27]. Medication costs were obtained from RED BOOK™ (www.ibm.com/products/micromedex-red-book) using wholesale acquisition costs, and were uniformly applied across sites of care. Diagnostic, transfusion, and procedure costs were taken from the Centers for Medicare & Medicaid Services Clinical Laboratory and Physician Fee Schedules (www.cms.gov/medicare/medicare) and the Hospital Outpatient Prospective Payment System (www.cms.gov/medicare/medicare-fee-for-service-payment/hospitaloutpatientpps). Unit costs were estimated by applying a payment-to-cost ratio to any Medicare reimbursement rates. Costs were adjusted according to the site of care where the HCRU occurred, to account for overhead costs; previous research has shown that cancer treatment costs are an average of 38% higher in hospital outpatient departments than in the physician’s office, and 35% higher in inpatient versus outpatient settings [28, 29].

Statistical Analysis

All results were analyzed descriptively, and no sample size or power calculations were conducted; HCRU data on all patients who were infused with ide-cel as part of the KarMMa trial (n = 128) were included in the analyses. Average total post-infusion costs per month were calculated among patients with ongoing status in that month. Three additional subgroup analyses were conducted to assess the costs of care: (1) for patients who received ide-cel in the US; (2) for patients who received the 450 × 106 CAR + T cell dose of ide-cel; and (3) for patients experiencing CRS/NT. A scenario analysis assuming a 7-day inpatient stay (no ICU admissions) rather than the 14-day required stay per trial protocol was also conducted. Patients censored due to the data cutoff date were excluded. All analyses were performed using Microsoft® Excel, R 4.0.2/RStudio 1.3.959, STATA 14 IC, or SAS 9.4.

Ethics Statement

This study does not contain any new analyses conducted by the authors with human participants or animals. The primary KarMMa study was conducted in accordance with the International Council for Harmonization guidelines for Good Clinical Practice and the principles of the Declaration of Helsinki. The study protocol was approved by local or independent institutional review boards or ethics committees at participating sites. All the patients provided written informed consent. The authors had permission to access and use the clinical trial database for this study. The efficacy and safety data from the KarMMa clinical trial along with the patient reported outcomes have been previously published [18, 22, 23].

Results

A total of 128 patients received ide-cel in the KarMMa trial and were included in this analysis. The median age was approximately 61 years, 59.8% were men (Table 1), and 81% were white. Patients had received a median of six prior antimyeloma treatment regimens, and 42% received the 450 × 106 CAR + T cell dose of ide-cel.

Table 1 Patient characteristics
Full size table

HCRU and Costs Through 24 Months Post-ide-cel Treatment

All patients had an inpatient stay according to the KarMMa trial protocol (14 days required), with a mean LOS of 22.3 days (Table 2). Approximately 20% (n = 26/128) of patients had an ICU stay, with a mean LOS of 1.6 days. The mean (SD) total LOS including both inpatient and ICU stays was 23.9 (14.4) days. Nearly all patients received antibiotics (96%, n = 123/128) and corticosteroids alone (45%, n = 58/128) or with tocilizumab (55%, n = 70/128) (Table 2). All medications were assessed post-infusion regardless of association with an AE or CRS/NT. Few patients had dialysis, intubation (used as a proxy for mechanical ventilation), or vasopressors. The total estimated costs over 24 months post-ide-cel infusion, not including the cost of ide-cel, were $115,614 per patient, driven primarily by facility costs (75%; $86,385). Over the 24-month post-infusion period, more than half of all costs were incurred in the first month post-ide-cel infusion (58%; $67,259) (Fig. 2). Facility costs accounted for 80% of Month 1 costs, likely attributable primarily to the 14-day inpatient stay for post-treatment monitoring required by the KarMMa trial protocol. Of note, the Month 1 facility costs accounted for 47% of all costs over the 2-year study period. Overall, nearly all costs were incurred during the first year after ide-cel infusion as compared with the second year (91%; Months 1–12, $105,491 vs. $10,122 for Months 13–24; Supplementary Material Fig. S1).

Table 2 HCRU through 24 months post-ide-cel infusion
Full size table
Fig. 2
figure 2

Monthly costs of care averages through 24 months post-ide-cel infusion. ide-cel idecabtagene vicleucel, USD United States dollar

Full size image

Subgroup and Scenario Analyses

Patients Receiving Ide-Cel in the US

A total of 94/128 (73%) patients who received ide-cel in the KarMMa trial were treated in the US. Characteristics of patients from the US were generally similar to those of the overall study population. The mean (SD) age was 60 (9.4) years, 57% were men, and 82% were white. Patients had received a mean of seven prior antimyeloma regimens.

Among patients treated in the US, the mean (SD) inpatient LOS was 20.6 (13.8) days (n = 94; 100%). Only 17 (18%, n = 17/94) US patients had an ICU stay, with a mean (SD) ICU LOS of 1.5 (4.4) days. The mean (SD) total LOS including both inpatient and ICU stays was 22.0 (14.8) days. As with the overall population, nearly all patients received antibiotics (95%, n = 89/94) and corticosteroids alone (46%, n = 43/94) or with tocilizumab (54%, n = 51/94). No patients had dialysis and few had intubation (4%, n = 4/94). The total estimated costs over 24 months post-ide-cel infusion were $109,500 per patient in the US, also driven by facility costs ($79,359; 72%) (Fig. 3).

Fig. 3
figure 3

Monthly costs of care averages among patients receiving ide-cel in the US. ide-cel idecabtagene vicleucel, US United States, USD US dollar

Full size image

Patients Receiving the 450 × 106 CAR + T Cell Dose of Ide-Cel

Of the 54/128 (42%) patients who received the 450 × 106 CAR + T cell dose of ide-cel, the mean (SD) inpatient LOS was 21.2 (12.5) days (100%, n = 54). Thirteen patients (24%, n = 13/54) had an ICU stay, with a mean (SD) ICU LOS of 1.9 (3.1) days. The mean (SD) total LOS including both inpatient and ICU stays was 23.1 (13.7) days. The total estimated costs over 24 months post-ide-cel infusion were $113,298 per patient, $2,316 lower than in the overall population (Fig. 4).

Fig. 4
figure 4

Monthly costs of care averages among patients receiving the 450 × 106 CAR + T cell dose of ide-cel. CAR chimeric antigen receptor, ide-cel idecabtagene vicleucel, USD United States dollar

Full size image

7 Day Post-Infusion Inpatient Stay

The scenario analysis that considered a 7-day inpatient stay after the ide-cel infusion (rather than the KarMMa trial protocol of 14 days), and assumed no ICU stays were required, presumed characteristics of patients identical to those of the overall study population. The 128 patients considered in the scenario had a mean (SD) inpatient LOS of 15.8 (13.3) days (100%, n = 128), and 18 patients (14%, n = 18/128) were assumed to have an ICU stay, with a mean (SD) ICU LOS of 1.2 (4.3) days. The mean (SD) total LOS including both inpatient and ICU stays was 16.9 (14.3) days. The scenario estimated a 24-month post-infusion cost of $92,294 per patient [$23,320 (20%) lower than with the 14-day inpatient stay] (Fig. 5). As with all other analyses, monthly costs were nominal after the first month and most costs occurred in the first year.

Fig. 5
figure 5

Monthly costs of care averages assuming a 7-day post-infusion standard inpatient stay. USD United States dollar

Full size image

Discussion

This study estimated the long-term costs of care following CAR T cell therapy for patients with TCE RRMM. Extrapolation of costs based on HCRU data from patients receiving single-infusion ide-cel treatment in the KarMMa trial showed substantially reduced HCRU and costs over 2 years after the initial treatment period. Overall, few patients required ICU stays, dialysis, intubation, or vasopressors, and nearly all costs were incurred during Year 1. Total costs of care in the first month after the ide-cel infusion accounted for more than half of all costs over 2 years, excluding the cost of ide-cel treatment. Facility costs accounted for the majority of Month 1 costs, and Month 1 facility costs accounted for nearly half of all costs over the 2-year study period. As such, the 2-year costs of care were 20% lower in the scenario analysis exploring the effect of a 7-day inpatient stay after the ide-cel infusion, with no ICU stay, as compared with the 14-day trial requirement and for those who received the 450 × 106 CAR + T cell dose of ide-cel. Taken together, these findings suggest a nominal, incremental monthly cost of care over 2 years after the ide-cel infusion, driven by the immediate treatment period costs.

Little data are available on the costs of care following CAR T cell therapy beyond those explored in this study. Jagannath et al. [15] analyzed a US administrative healthcare claims database to estimate HCRU and costs incurred by patients with TCE RRMM and found mean RRMM-related healthcare costs of $655,524 during a mean 20.9 months of follow-up. Madduri et al. [16] analyzed treatment patterns and healthcare costs among patients with TCE RRMM in the US and estimated mean monthly MM-related costs of $35,657 per patient. Ghanem and Shi [30] recently estimated annual costs of ide-cel or ciltacabtagene autoleucel for patients with RRMM in the US. They estimated annual administration and AE management costs (inflated to 2022 costs) of $108,520 and $100,535 for ide-cel and ciltacabtagene autoleucel, respectively, with drug acquisition costs of $419,500 and $465,000 [30]. Our cost estimates suggest that annual costs are very different year on year, as the 24-month costs were only about 10% higher than Year 1 costs. Our sensitivity analysis exploring a likely more realistic inpatient stay of 7 rather than 14 days reduced 2-year costs by a further 20%, and assumed no ICU stays, which was generally aligned with the low ICU utilization observed among patients receiving ide-cel in the KarMMa trial.

These findings should be interpreted in the context of some important considerations. This analysis was based on patients who received ide-cel in the KarMMa clinical trial; thus, findings may not be generalizable to other CAR T cell therapies or to settings outside of a clinical trial. As such, cost estimates in this study were only representative of patients with available data from the KarMMa trial, and the numbers of patients contributing to each monthly cost estimate decreased over 24 months as patients were lost to follow-up. It should also be noted that costs represented national averages and may not be generalizable to specific institutions or health systems. Some costs were based on Medicare rates which would differ for patients with other reimbursement schemes, such as commercially insured patients in the US, although most patients with RRMM in the US would likely have primary insurance coverage through the Medicare program. These were extrapolated costs based on observed HCRU in the KarMMa trial. Data on true costs for each site of care were not available, and unit costs may differ across countries and payers. HCRU was captured post-infusion through a 24-month period, but the attribution of reason for the HCRU use was not specified and some HCRU captured may be related to monitoring or management care of the patient for reasons not related to the ide-cel infusion or management of treatment-emergent AEs (i.e., all medications taken post-infusion, including corticosteroids, were captured in HCRU costs whether or not they were associated with an AE or CRS/NT). Right-censoring due to death, discontinuation, or data cutoff included 2-year costs estimated from available data of uncensored patients. Unobserved variables for censored patients may have contributed to other unseen costs. Finally, the 14-day inpatient stay required by the KarMMa trial protocol may not reflect real-world practice and may have inflated the estimated facility costs, which were the largest driver of all costs of care. This was illustrated in the sensitivity analysis of a 7-day inpatient stay and no ICU stay that showed lower costs overall. While other uncertainties or practices not reflecting the real-world use of ide-cel may exist in areas like medications, diagnostics, and procedures, a reduction in inpatient stay with no ICU use was the single scenario seen as most likely to better reflect real-world use. In addition, this change principally impacted the facility category of HCRU, which made up close to three-quarters of the mean costs assessed.

Conclusion

This analysis of HCRU and costs following ide-cel treatment showed that few patients in the KarMMa trial required ICU stays or other potentially costly HCRU as follow-up care. Nearly half of all costs over the 24 months after the ide-cel infusion were attributable to facility costs during the initial treatment period, followed by nominal incremental medical care costs over the subsequent 23 months. Subgroup and scenario analyses suggested that costs of care may be lower in less rigid circumstances than the clinical trial setting, and further research is needed to illustrate the real-world costs of follow-up care following CAR T cell therapy for patients with TCE RRMM in clinical practice.