The patient population was based on the CARTITUDE-1 trial (registered at ClinicalTrials.gov as NCT03548207), a phase IB/II clinical trial evaluating the efficacy and safety of cilta-cel CAR-T among patients with RRMM, who received cilta-cel CAR-T infusion, i.e., a modified intention-to-treat (mITT) population. A total of 113 patients with RRMM with prior exposure to a PI, an IMiD, and an anti-CD38 monoclonal antibody underwent apheresis [intention-to-treat (ITT) population] and 97 (85.8%) received cilta-cel CAR-T therapy. Sixteen patients with RRMM underwent apheresis but did not ultimately receive CAR-T therapy. Primary reasons for not receiving therapy included pre-infusion mortality, disease progression, and patient/physician decision [3]. Micromedex RED BOOK, a commercially licensed database [11] and the CMS 2021 Physician Fee Schedule, a publicly available database [10], were used to calculate costs in this study. No patient level data were used in the analyses and no institutional review board (IRB) approval was needed.

Costs are presented in 2021 US dollars (USD) from the perspective of a commercial payer with Medicare-eligible members and are limited to direct medical costs. Cost components were identified for three phases of CAR-T therapy: pre-, peri-, and postinfusion (Fig. 1). Management of adverse events (AEs) associated with cilta-cel infusion were considered as a separate cost component. US prescribing information (USPI) for cilta-cel [1], clinical trial data [7], publicly-available medical cost databases (discussed in detail below) [8,9,10,11], and published literature [12] were used to identify the cost components and unit costs.

Costs components by stage of the CAR-T therapy beginning with pre-infusion procedures

To address gaps in the literature surrounding healthcare resource utilization (HCRU) in the treatment of RRMM with CAR-T, a double blinded Delphi panel of clinical experts was convened [13, 14]. One of the primary objectives of the panel (Study Delphi Panel) was to gain consensus on HCRU associated with CAR-T therapy for RRMM. The method used a two-phase process, following established Delphi methods [13, 14]. The details of the methodology are described in the supplementary materials (Supplementary Materials Table 1).

Pre-infusion costs included apheresis, bridging therapy, and conditioning chemotherapy costs (Table 1). We assumed outpatient site of care for all pre-infusion procedures. The cost of apheresis was sourced from the CMS 2021 Physician Fee Schedule, a publicly available database [10].

There is no single standard of care for bridging therapy, with chosen treatments generally based on physician choice and specific to patient needs [2]. In our base case analysis, the proportion of patients receiving bridging therapy, 75.26%, is based on the CARTITUDE-1 trial. In all analyses, bridging therapy costs were accrued for 3 weeks based on clinician expert opinion from the Study Delphi Panel. Weekly bridging therapy costs were based on the average of ten non-CAR-T RRMM regimens weighted by market share [11].

It was assumed all patients received lymphodepleting chemotherapy in an outpatient setting as a conditioning therapy. Conditioning therapy comprised 30 mg/m2 fludarabine and 300 mg/m2 cyclophosphamide infusions administered for 3 days based on the cilta-cel clinical trial protocol [7], consistent with the USPI for CARVYKTI® [1]. Baseline surface area (BSA) was assumed to be 1.87 m2 [15].

The acquisition costs of bridging therapy and conditioning chemotherapy were the wholesale acquisition cost prices sourced from Micromedex RED BOOK, a commercially licensed database [11]. Infusion costs were sourced from the CMS 2021 Physician Fee Schedule [10].

Peri-infusion costs included inpatient hospital days and outpatient visits (Table 1). The primary analysis assumed 100% inpatient administration of cilta-cel. Average length of inpatient stay was assumed to be 7 days, and any extended inpatient stays resulting from CRS are accounted for in AE costs, which are separately costed. In CARTITUDE-1, the median number of days to onset of CRS symptoms in patients who received a cilta-cel CAR-T infusion and subsequently developed CRS was 7 days [1].

Scenario analyses were also conducted assessing 85%/15% and 70%/30% inpatient/outpatient cilta-cel infusion administration. For the outpatient dosing in each scenario, it was assumed patients incurred 1 day of hospitalization costs for the day of the infusion. Additional hospitalization time and costs would be associated with AEs (discussed below). The number of outpatient visits immediately following infusion in both administration settings (7 visits in the inpatient setting and 11 visits in the outpatient setting; Table 1) were based on clinician expert opinion from the Study Delphi Panel.

Cost per inpatient day was sourced from an Agency for Healthcare Resource and Quality (AHRQ) statistical brief [9], as cited in a published evaluation of CAR-T therapies [16], and inflated to 2021 values using the medical care component of the Consumer Price Index (CPI) from the US Bureau of Labor Statistics (BLS) [17]. Cost per outpatient visit was from the CMS physician fee schedule, the standard source use for outpatient costs [10].

Postinfusion costs included 100-day postinfusion monitoring costs and additional infusion monitoring costs required in the remainder of the first year (Table 1). Utilization of 100-day and subsequent first-year monitoring resources was based on clinician expert opinion from the Study Delphi Panel. Briefly, the analysis assumed 100-day monitoring comprised twice weekly outpatient visits, laboratory testing frequency of two to three times per week for the first month followed by less frequent testing, and one bone marrow biopsy. Subsequent first-year monitoring comprised once-monthly outpatient visits and laboratory testing. Monitoring resource costs were sourced from the CMS 2021 physician fee schedule [10]. Resource use and unit costs are detailed in Supplementary Materials Table 2. Postinfusion monitoring cost components with unit costs and resource utilization are also presented in Supplementary Materials Table 2.

Adverse event costs included costs of managing AEs of interest associated with cilta-cel therapy, including hospitalization when required (Table 2). All grades of cytokine release syndrome (CRS) and neurologic toxicity AEs were included in the analysis. Additional AEs of interest included in the analysis were grade ≥ 3 AEs occurring in > 5% of patients receiving a cilta-cel infusion sourced from the cilta-cel USPI, which reflected a median duration of follow-up of 18 months [1]. CRS and neurologic toxicity AE costs were derived from a published study, using the assumption that grade ≥ 3 CRS events with hemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH) had similar costs to grade 4 CRS [12]. All other AE costs were retrieved from the AHRQ Healthcare Cost and Utilization Project website [8]. All costs were inflated to 2021 values using the medical care component of the US BLS CPI [17].

Two sets of scenarios were explored in addition to the scenarios varying cilta-cel administration setting. In scenario A, the percent of patients receiving bridging therapy was increased to 100%. All other assumptions were identical to the base case.

Currently, there are two CAR-T therapies (cilta-cel and ide-cel) approved for patients with RRMM, and these two clinical trials reported different bridging therapy rates [7, 18]. In scenario B, we estimated the impact of changing the bridging therapy rate on cilta-cel administration cost by increasing the rate from 75.26% in the cilta-cel trial to 82.20%, the weighted average of the bridging therapy use rates from the two CAR-T clinical trials [7, 18].

The impact of 15% and 30% of patients receiving outpatient infusions were also explored in scenarios A and B.