Abstract
BACKGROUND: Despite recent advancements in the therapeutic landscape, multiple myeloma (MM) remains incurable. There are multiple treatment options available with a novel mechanism of action, but there is limited evidence describing the economic burden among patients with MM exposed to different drug classes and combinations and across different health care settings.
OBJECTIVE: To describe all-cause and MM-related health care resource utilization (HCRU) and costs among patients with MM exposed to different drug classes and combinations (ie, double-class and triple-class–exposed) and characterize the economic burden in different health care settings among these patients with MM.
METHODS: We conducted a retrospective cohort study using the IBM MarketScan databases. The study included adult patients (aged ≥18 years) diagnosed with MM between December 1, 2015, and December 31, 2019. The study sample comprised double-class–exposed (DCE) and triple-class–exposed (TCE) cohorts, categorized based on their earliest exposure to different combinations of immunomodulatory drugs, proteasome inhibitors, or targeted monoclonal antibody. Patients with at least 1 subsequent line of therapy following the categorization were included, and the start date of the first subsequent line of therapy was the index date. The primary outcomes were all-cause and MM-related HCRU and costs during the follow-up period. Costs were stratified across 8 care settings defined by place of service. The Kaplan-Meier sample average technique was used to estimate the cumulative mean outcomes, accounting for differential follow-up periods. The outcomes were reported as per patient per month (PPPM). 18 years) diagnosed with MM between December 1, 2015, and December 31, 2019. The study sample comprised double-class–exposed (DCE) and triple-class–exposed (TCE) cohorts, categorized based on their earliest exposure to different combinations of immunomodulatory drugs, proteasome inhibitors, or targeted monoclonal antibody. Patients with at least 1 subsequent line of therapy following the categorization were included, and the start date of the first subsequent line of therapy was the index date. The primary outcomes were all-cause and MM-related HCRU and costs during the follow-up period. Costs were stratified across 8 care settings defined by place of service. The Kaplan-Meier sample average technique was used to estimate the cumulative mean outcomes, accounting for differential follow-up periods. The outcomes were reported as per patient per month (PPPM).
RESULTS: The study included 1,521 patients with MM, of whom 1,016 (66.8%) were DCE and 505 (33.2%) were TCE. The mean total all-cause health care costs were $20,338 PPPM, and approximately 85% of the total all-cause costs were MM-related. The mean all-cause and MM-related total costs were driven by overall drug costs primarily attributed to MM treatment and administration costs. The TCE cohort was associated with more HCRU and incurred higher costs than the DCE cohort across all categories. The hospital-based ambulatory setting had the highest all-cause and MM-related costs during the follow-up period: $7,302 (95% CI = $6,801-$7,784) PPPM and $6,695 (95% CI = $6,239-$7,136) PPPM, respectively.
CONCLUSIONS: The study findings suggest that the economic burden following exposure to multiple drug classes and combinations is substantial, especially among the TCE cohort and in the ambulatory setting. These findings highlight the need for more effective treatments that can mitigate the economic burden of patients with MM. Future research on the HCRU and costs related to recently approved MM treatments with novel mechanisms is warranted.
DISCLOSURES: At the time of this study, Dr Yang was a postdoctoral fellow and the fellowship was supported by GSK. Dr Boytsov is a full-time employee of GSK. Dr Carlson discloses consulting fees from Pfizer, AbbVie, and Genentech. Dr Barthold reports no disclosures.
Plain language summary
This study examined health care use and costs in people with multiple myeloma (MM) who have received multiple drug classes and combinations. We observed high monthly costs and health care use among these patients, especially in patients exposed to 3 drug classes compared with those exposed to 2 drug classes. The findings from this study highlight the need for more effective treatments to treat MM and reduce the health care burden.
Implications for managed care pharmacy
This study’s key findings are aligned with previous studies that have estimated high monthly costs among patients with MM following multiple drug class exposure, especially triple-class–exposed patients. This study provides additional evidence of substantial economic burden among patients with MM and highlights the differences in economic burden across class-exposure status and health care settings. Thus, the study findings can inform decisions to optimize resource allocations, including access to existing and future MM therapies with novel mechanisms.
Multiple myeloma (MM) is a rare malignancy of plasma cells characterized by the proliferation of malignant cells in the bone marrow, which results in immune suppression.1 Common complications include hypercalcemia, renal dysfunction, anemia, bone loss, and bone fractures.1 In the United States, MM is the third most common hematologic malignancy and is most frequently diagnosed among people aged 65 to 74 years.2 In 2023, it is estimated that there will be approximately 35,730 new cases of MM and 12,640 patients will die of this disease in the United States.2
In the last 2 decades, several classes of drugs have emerged and significantly improved patients’ clinical outcomes such as immunomodulatory drugs (IMiDs), proteasome inhibitors (PIs), and targeted monoclonal antibodies (mAbs).3 These drug classes enabled patients to live longer, and the 5-year relative survival rate increased from 31.6% in the 1990s to 57.9% in 2012-2018 in the United States.2 The National Comprehensive Cancer Network Clinical Practice Guidelines in Multiple Myeloma (NCCN Guidelines) recommend that the initial therapy should involve combinations of an IMiD, a PI, and/or an mAb with corticosteroids (eg, dexamethasone).4 The standard of care is a combination of 3 drug classes; however, a combination of 2 drug classes is recommended for patients who are frail.4 Following the initial treatment, the guideline recommendations for the subsequent therapies vary, and several factors influence the selection, such as the patient’s response to previous treatment and tolerance to toxicities.4 Despite recent advancements in the therapeutic landscape, MM remains incurable, and most patients eventually relapse or become refractory to classes of drugs, requiring several lines of therapies (LOTs).3
Despite previous exposures or refractoriness, patients with MM who have received multiple prior LOTs are frequently re-treated with the same classes of drugs and are more likely to experience poor outcomes.5 Most patients with MM relapse and receive multiple salvage therapies, which decreases patients’ response to treatments while the burden of MM-related symptoms and complications increases.6 In addition, previous studies have documented that MM is associated with a substantial financial burden, which is more significant among patients with disease progressions and for those who require several LOTs.7-10 Although there are several available treatment options for these patients, the treatment efficacy diminishes with increased exposure, which has prompted the development of new therapies with novel mechanisms that can potentially delay disease progression and improve clinical outcomes. The US Food and Drug Administration (FDA) has recently approved several therapies, including antibody-drug conjugates (ADC), chimeric antigen receptor T-cell therapy (CAR-T), and bispecific T-cell engager therapy.11 These therapies are indicated as monotherapy in patients who have been heavily pretreated with IMiDs, PIs, and/or mAbs.
Although recent studies have described the health care resource utilization (HCRU) and costs among newly diagnosed and heavily pretreated patients with MM, there is limited evidence describing the economic burden among patients with MM exposed to different drug classes and combinations and across different health care settings.7-10,12-14 A deeper and more granular assessment of the HCRU and costs associated with the patients with MM who can potentially benefit from the novel therapies would aid in decisions to optimize resource allocations, including access to existing and future therapies.
This study aimed to describe all-cause and MM-related HCRU and costs among patients with MM exposed to different drug classes and combinations (eg, double-class–exposed [DCE] and triple-class–exposed [TCE]) from a third-party payer perspective in the United States. The study also aimed to describe and characterize all-cause and MM-related health care costs incurred in different health care settings among these patients with MM.
Methods
STUDY DESIGN AND DATA SOURCE
This was a retrospective observational cohort study using de-identified administrative claims from the IBM MarketScan Commercial Claims and Encounters (CCAE) and Medicare Supplemental (MDCR) databases between December 1, 2015, and December 31, 2019 (Figure 1). We chose December 1, 2015, as the starting date because it was the earliest date we can observe claims for daratumumab and elotuzumab (mAb) in MarketScan following their FDA approvals in November 2015. The databases contain fully paid and adjudicated inpatient, outpatient, and pharmacy insurance claims data of active employees, their spouses, and dependents covered by employer-sponsored private health insurance in the United States and Medicare-eligible retirees covered by Medicare Advantage and Medicare Supplemental health insurance plans.15 The MarketScan databases are fully compliant with the Health Insurance Portability and Accountability Act of 1996.15
FIGURE 1.
Study Schematic
STUDY POPULATION AND SAMPLE
This study included adult patients (aged ≥ 18 years) with at least 1 confirmed diagnosis of MM, defined by at least 1 inpatient service claim or at least 2 outpatient service claims 30 to 365 days apart with a primary or secondary diagnosis of MM and who have completed at least 1 cycle of IMiDs, PIs, or mAbs during the study period. Patients with at least 1 month of follow-up were included in this study. All International Classification of Diseases, Ninth Revision (ICD-9) and International Classification of Diseases, Tenth Revision (ICD-10) codes used for inclusion and exclusion and MM treatments are shown in Supplementary Table 1 (available in online article).
From the overall study population, the study sample comprised 2 cohorts categorized based on their earliest exposures to different combinations of IMiDs, PIs, and mAbs through monotherapy or combination regimens. The DCE cohort consisted of patients exposed to the following combinations of classes: IMiD and PI, IMiD and mAb, or PI and mAb. The TCE cohort consisted of patients exposed to all 3 drug classes of interest. Because the NCCN Guidelines include corticosteroids (ie, dexamethasone) in their recommended combination regimens for MM, evidence of receiving corticosteroids did not count as a drug class in the definitions for DCE and TCE.
The patients’ exposures to different drug classes and combinations were defined by completing at least 1 LOT that contained IMiDs, PIs, or mAbs, regardless of the sequences or combinations used. The earliest LOT that categorized patients as DCE or TCE was the “pre-index LOT.” Patients with at least 1 subsequent LOT following the categorization were included, and the first subsequent LOT was the “index LOT” with the start date as the index date (Figure 1). The study included patients with continuous enrollment during the baseline period, defined as 6 months prior to the index date. The study outcomes were assessed during the follow-up period, defined as the time from the index date to the end of the study period or the end of continuous enrollment, whichever occurred first (Figure 1).
LINE OF THERAPY
The previously recommended guideline for determining the number of prior LOTs in MM was adapted and used in this study to define a LOT.16 The guideline also described the conditions for a new LOT to identify the “index LOT” following the “pre-index LOT.”16 In general, a LOT was defined as at least 1 complete cycle of a single drug, a regimen that consists of combinations of different drugs, or planned sequential treatments of various regimens. A treatment was considered a new LOT if one of the following conditions was met: (1) a new MM treatment was added after the initial 28 days after the start of a regimen and (2) discontinuation of all treatments in a regimen and then restart the regimen with at least 1 other regimen administered in between.
However, the following scenarios did not meet the conditions for a new LOT: (1) discontinuation of at least 1 MM treatment in a regimen, but not all treatments in the regimen; (2) discontinuation and restart of all MM treatments in a regimen without any new treatments administered in between; and (3) discontinuation of at least 1 MM treatment in a regimen, but not all, and then restart of the discontinued treatment.
Identifying a LOT in retrospective claims is complex and challenging, as the data may not always provide sufficient detail on treatment use. The conditions described in the guideline are vital to ensure that only significant treatment changes are classified as new LOTs. This allows for a more precise and reliable estimation of patients’ exposures to different drug classes and combinations, and outcomes following the exposures.
STUDY MEASURES AND OUTCOMES
Baseline demographic and clinical characteristics were assessed for each patient during the baseline period. Information regarding sex, age, Charlson Comorbidity Index (CCI) scores, geographic regions, insurance type (eg, commercial or Medicare supplemental), and year of the index date were reported.
All-cause and MM-related HCRU and costs for each patient were assessed during the baseline and follow-up periods. MM-related HCRU and costs were defined by claims with ICD-9 and ICD-10 codes for MM. Total costs were identified using the total payment variable within the MarketScan database. All costs were adjusted to 2021 US dollars using the medical care component of the Consumer Price Index for urban consumers.17
All-cause and MM-related HCRU measures included the number of emergency department (ED) visits, outpatient physician office visits, other outpatient visits (ie, all outpatient visits excluding ED and outpatient physician office visits), outpatient pharmacy prescriptions, number of hospitalizations, and days of inpatient stays. All-cause health care costs were ED, outpatient, inpatient, and overall drug, and the sum of these 4 categories is referred to as total health care costs. MM-related health care costs were MM treatment, MM treatment administration, ED, outpatient, and inpatient, and the sum of these 5 categories is referred to as total MM-related costs. All-cause and MM-related HCRU and costs among patients with at least 1 ED, outpatient, or inpatient visit were also reported.
Different health care settings were defined by the place of service code (ie, MarketScan variable STDPLAC).18 Based on the place of service codes present in all claims for the study sample, the codes were sorted and categorized into 8 different care settings (Supplementary Table 2). The categorization of each code was adapted from the previous literature that had classified the place of service code to identify settings for prescriptions.19 The total all-cause and MM-related costs incurred by patients in each setting during the follow-up period were reported.
Subgroup analyses were conducted based on patient characteristics and health care cost types. The subgroups were compared by sex, age, calendar year of the index date, exposure to mAbs, and insurance type. Age was stratified into those aged younger than 65 years and those aged 65 years or older. The health care costs were stratified into patient out-of-pocket costs (ie, the sum of coinsurance, copays, and deductibles) and payer costs (ie, identified using the net payment variable within the MarketScan database). All-cause and MM-related total costs incurred by patients in each subgroup and by each cost type during the follow-up period were reported.
STATISTICAL ANALYSIS
All analyses conducted in this study were descriptive, and no hypothesis testing was conducted. Continuous variables were summarized with mean and SD or 95% CIs, and categorical variables were summarized with frequencies and percentages. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc.).
The cumulative HCRU and costs during the follow-up period were estimated using the Kaplan-Meier sample average (KMSA) technique to account for differential follow-up time due to censored patients. Patients were censored at the event of discontinuation in insurance coverage or the end of the study period, whichever occurred first. The KMSA technique estimated the cumulative mean outcomes during the follow-up period by calculating the sum of the products of the monthly probability of patients remaining in the health plan and the monthly mean outcomes (ie, HCRU and costs) among patients who remain in the health plan.20,21 The survival weights were calculated with a follow-up time of 48 months. Nonparametric bootstrapping methods with 1,000 independent samples were used to generate the 95% CIs for the KMSA estimates. The costs and HCRU were reported as per patient per month (PPPM).
Results
The study included a total of 1,521 patients with MM (1,016 [66.8%] were DCE and 505 [33.2%] were TCE) (Supplementary Figure 1). Table 1 shows the baseline patient characteristics. The mean age of the overall cohort at the index date was 62.3 years, and 57% were male. The majority of the patients were under 65 years (68.6%) and enrolled in commercial insurance during the study period (67.7%). The proportion of the overall cohort with TCE, rather than DCE, increased across 4 years of the study period. The majority of the DCE cohort consisted of patients exposed to combinations of an IMiD and a PI (91.7%), which are the most common drug class combinations recommended by the NCCN Guidelines (Table 1).4
TABLE 1.
Baseline Patient Demographic and Clinical Characteristics
| Overall(N = 1,521) | Double-class–exposed(n = 1,016) | Triple-class–exposed(n = 505) | |
|---|---|---|---|
| Sex, n (%) | |||
| Male | 867 (57) | 576 (56.7) | 291 (57.6) |
| Female | 654 (43) | 440 (43.3) | 214 (42.4) |
| Age, mean (SD), yearsa | 62.3 (10.5) | 62.5 (10.8) | 62.1 (9.8) |
| Age group in years, n (%) | |||
| < 65 | 1,044 (68.6) | 684 (67.3) | 360 (71.3) |
| 65 + | 477 (31.4) | 332 (32.7) | 145 (28.7) |
| CCI group, n (%) | |||
| 0 | 0 | 0 | 0 |
| 1 | 0 | 0 | 0 |
| 2 | 521 (34.3) | 368 (36.2) | 153 (30.3) |
| 3 + | 1,000 (65.7) | 648 (63.8) | 352 (69.7) |
| US region of residence, n (%) | |||
| North central | 380 (25) | 250 (24.6) | 130 (25.7) |
| Northeast | 348 (22.9) | 243 (23.9) | 105 (20.8) |
| South | 590 (38.8) | 396 (39) | 194 (38.4) |
| West | 202 (13.2) | 126 (12.4) | 76 (15.1) |
| Unknown | 1 (0.1) | 1 (0.1) | 0 |
| Insurance type | |||
| Commercial | 1,030 (67.7) | 670 (65.9) | 360 (71.3) |
| Medicare | 491 (32.3) | 346 (34.1) | 145 (28.7) |
| Year of index LOT initiation, n (%) | |||
| 2016 | 305 (20) | 303 (29.8) | 2 (0.4) |
| 2017 | 406 (26.7) | 255 (25.1) | 151 (29.9) |
| 2018 | 401 (26.4) | 233 (22.9) | 168 (33.3) |
| 2019 | 409 (26.9) | 225 (22.2) | 184 (36.4) |
| Previous class exposures, n (%) | |||
| Double-class–exposed | 1,016 (66.8) | ||
| IMiD + PI | 932 | ||
| IMiD + mAb | 44 | ||
| PI + mAb | 40 | ||
| Triple-class–exposed | 505 (33.2) | ||
a Age as of the index date.
CCI = Charlson-Deyo Comorbidity Index; IMiD = immunomodulatory drug; LOT = line of therapy; mAb = targeted monoclonal antibodies; MM = multiple myeloma; PI = proteasome inhibitors.
During the baseline period, the overall cohort had 5.20 (95% CI = 5.00-5.40) PPPM all-cause physician office visits, 4.83 (95% CI = 4.69-4.96) PPPM outpatient prescriptions, and 0.11 (95% CI = 0.10-0.12) PPPM hospitalizations with an average length of stay of 1.02 days (95% CI = 0.91-1.12) PPPM (Supplementary Table 3). The majority of the HCRU were MM-related. The mean total all-cause health care costs during the baseline period were $29,849 (95% CI = $28,673-$31,026) PPPM, and most of the costs (81%) were MM-related. When stratified by class-exposure status, the TCE cohort incurred higher mean total all-cause and MM-related costs and HCRU compared with the DCE cohort (Supplementary Table 3).
Following the index date, the mean duration of follow-up for the overall, DCE, and TCE cohorts were 16.3 months, 18.4 months, and 12.1 months, respectively (Table 2). The overall cohort’s mean number of all-cause physician office visits, outpatient prescriptions, and hospitalizations during the follow-up period were 3.03 (95% CI = 2.91-3.13), 3.15 (95% CI = 3.06-3.24), and 0.07 (95% CI = 0.06-0.07) PPPM, respectively. A majority of HCRU was MM-related. MM-related physician office visits and other outpatient visits were 2.21 (95% CI = 2.14-2.29) PPPM and 2.49 (95% CI = 2.42-2.60) PPPM (Table 2). The majority of all-cause hospitalization was MM-related, 0.06 (95% CI = 0.06-0.07) PPPM, with an average length of stay of 0.61 days (95% CI = 0.56-0.66) PPPM. The proportion of the overall cohort with at least 1 MM-related ED, inpatient, and outpatient visit were 22%, 50%, and 99%, respectively (Supplementary Table 4). When we stratified HCRU by class-exposure status, the TCE cohort had higher utilization across all categories for all-cause and MM-related HCRU than the DCE cohort. Particularly, the TCE cohort had a higher number of all-cause and MM-related outpatient visits despite the shorter follow-up compared with the DCE cohort (Table 2).
TABLE 2.
All-Cause and MM-Related HCRU Incurred During the Follow-Up Period
| Overall (N = 1,521) | Double-class–exposed (n = 1,016) | Triple-class–exposed (n = 505) | |
|---|---|---|---|
| Follow-up duration in months, mean (SD) | 16.31 (13.20) | 18.40 (13.91) | 12.10 (9.51) |
| All-cause HCRU, PPPM, mean (95% CI) | |||
| ED visits | 0.06 (0.05-0.06) | 0.05 (0.04-0.05) | 0.08 (0.08-0.09) |
| Outpatient | |||
| Physician office visits | 3.03 (2.91-3.13) | 2.68 (2.60-2.76) | 4.05 (3.98-4.13) |
| Other outpatient visits | 3.56 (3.42-3.73) | 3.00 (2.91-3.09) | 5.19 (5.09-5.26) |
| Outpatient prescriptions | 3.15 (3.06-3.24) | 2.92 (2.83-2.99) | 3.84 (3.79-3.89) |
| Hospitalizations | 0.07 (0.06-0.07) | 0.05 (0.05-0.05) | 0.11 (0.11-0.12) |
| Inpatient length of stay, days | 0.63 (0.58-0.73) | 0.53 (0.49-0.56) | 0.96 (0.91-0.99) |
| MM-related HCRU, PPPM, mean (95% CI) | |||
| ED visits | 0.02 (0.02-0.03) | 0.02 (0.01-0.02) | 0.04 (0.04-0.05) |
| Outpatient | |||
| Physician office visits | 2.21 (2.14-2.29) | 1.93 (1.87-1.99) | 3.07 (3.01-3.13) |
| Other outpatient visits | 2.49 (2.42-2.60) | 2.06 (1.99-2.12) | 3.77 (3.72-3.82) |
| Outpatient prescriptions | 0.65 (0.62-0.67) | 0.64 (0.62-0.66) | 0.68 (0.66-0.69) |
| Hospitalizations | 0.06 (0.06-0.07) | 0.05 (0.04-0.05) | 0.11 (0.11-0.11) |
| Inpatient length of stay, days | 0.61 (0.56-0.66) | 0.50 (0.47-0.54) | 0.93 (0.89-0.97) |
ED = emergency department; HCRU = health care resource utilization; MM = multiple myeloma; PPPM = per patient per month.
The overall cohorts’ mean total all-cause health care costs were $20,338 (95% CI = $19,384-$20,956) PPPM, and approximately 85% of all-cause total health care costs were MM-related (Figure 2). The mean total all-cause and MM-related costs were driven by overall drug costs, primarily attributed to MM treatment and administration costs (Figures 2 and 3). When stratified by class-exposure status, the mean total all-cause health care costs for the DCE and the TCE cohorts were $17,171 (95% CI = $16,458-$17,599) PPPM and $29,791 (95% CI = $28,804-$30,020) PPPM, respectively. The TCE cohort incurred higher all-cause and MM-related costs across all categories compared with the DCE cohort (Figures 2 and 3).
FIGURE 2.
Components of the Total All-Cause Costs for the Overall, Double-Class–Exposed, and Triple-Class–Exposed Cohorts During the Follow-Up Period
FIGURE 3.
Components of the Total MM-Related Costs for the Overall, Double-Class–Exposed, and Triple-Class–Exposed Cohorts During the Follow-Up Period
Among 8 different health care settings, hospital-based ambulatory encounters had the highest costs during the follow-up period with mean total all-cause and MM-related costs of $7,302 (95% CI = $6,801-$7,784) PPPM and $6,695 (95% CI = $6,239-$7,136) PPPM, respectively. The inpatient hospital and office-based ambulatory were the next highest cost settings (Supplementary Figure 2).
We examined subgroups based on health care cost types and patient characteristics (Supplementary Table 5). A majority of the health care costs were borne by payers, with total all-cause costs of $16,129 (95% CI = $15,512-$16,771) PPPM. The mean total all-cause health care costs were similar in male patients and female patients. Patients aged younger than 65 years incurred higher total all-cause costs compared with patients aged 65 years or older. Similarly, patients with commercial insurance incurred higher total all-cause costs compared with patients with Medicare ($22,693 [95% CI = $22,138-$23,403] and $15,142 [95% CI = $14,732-$15,458], respectively). In addition, we observed an increasing trend in the mean total all-cause and MM-related costs across years of the index date. Patients exposed to an mAb had higher mean total all-cause and MM-related costs than patients unexposed to the drug class. We observed the same trend in the mean total MM-related health care costs for these subgroups (Supplementary Table 5).
Discussion
This retrospective observational cohort study assessed all-cause and MM-related HCRU and costs among patients exposed to different drug classes and combinations and costs incurred in different health care settings. Using the KMSA method to examine MarketScan data, we found that the mean total all-cause health care costs were $20,338 PPPM, and approximately 85% of the total all-cause costs were MM-related. Overall drug costs were the primary cost driver, followed by outpatient, inpatient, and ED costs. When stratified by the class-exposure status, the TCE cohort was associated with higher HCRU and incurred higher costs compared with the DCE cohort ($29,791 vs $17,171 PPPM). Among different health care settings, the mean total all-cause and MM-related health care costs for the hospital-based ambulatory setting were the highest, followed by inpatient and office-based ambulatory settings.
A prior study has shown that the survival outcomes remain poor among patients with MM who have been exposed to multiple drug classes.22 In addition to poor survival outcomes, the findings from this study provide deep and granular evidence of the economic burden of patients with MM following their exposure to IMiDs, PIs, and/or mAbs. To our knowledge, this was the first study to describe the economic burden among DCE patients with MM and across different health care settings. The results of this study highlighted high monthly costs incurred by patients with MM and differences in economic burden across class-exposure status, patient characteristics, and different health care settings. Variation in costs across patients could be caused by factors including variation in treatment intensity at different ages as well as variation in treatment outcomes associated with different therapies. These findings emphasize that future therapies with novel mechanisms that improve health outcomes and reduce economic burden will be valuable to these patients with MM.
The findings of our study were similar to previously published literature that evaluated the economic burden among patients with MM following drug class exposures in the United States. The studies by Madduri et al and Jagannath et al both described the costs among TCE patients using the MarketScan CCAE and MDCR databases. There was some overlap in study periods across the 3 studies where Madduri et al included patients from December 1, 2015, to September 30, 2018, and Jagganath et al included patients from January 1, 2009, to February 28, 2021. Madduri et al and Jagganath et al found mean total all-cause health care costs of $37,033 PPPM and $34,578 PPPM, respectively.12,13 The sample size of the TCE cohort in our study was larger than the 2 studies, which resulted from less restrictive inclusion criteria, a different study period, and different conditions that define a LOT and a new LOT. Despite differences in sample populations, all 3 studies reported that most of the total all-cause health care costs were MM-related, and MM therapy and administration costs represented more than 50% of the mean total MM-related health care costs.12,13
We assessed the mean total all-cause and MM-related health care costs incurred in different health care settings. Hospital-based ambulatory encounters incurred the highest mean total all-cause and MM-related health care costs. Several reasons may explain this. First, most MM treatments are administered as injections, and a hospital-based ambulatory setting is where patients with MM most commonly receive treatments, which are more costly than oral treatments. Second, the use of regimens containing more recently approved novel treatments (eg, carfilzomib, daratumumab, and elotuzumab) have increased in recent years because they have demonstrated improved clinical outcomes when administered in combination with older treatments (eg, lenalidomide and bortezomib).23 In addition, Hollmann et al estimated high average monthly costs of receiving triplet regimens containing these newer treatments ranging from $13,890 to $27,432 per patient.24 Thus, the high costs incurred in the hospital-based ambulatory setting are primarily attributable to the high costs associated with regimens that contain newer novel treatments.
In the baseline patient demographic and clinical characteristics of our study sample, we observed an increasing proportion of TCE, rather than DCE, across years of the study. Similarly, Braulin et al reported a trend of higher use of the triplet regimens across all LOTs over the study period.25 Furthermore, we observed an increasing trend in the mean total all-cause and MM-related health care costs across years of the study. Thus, a continuous increase in the number of TCE patients and total health care costs suggests that the economic burden among patients with MM will also continue to increase, which further highlights the need for future therapies that offer both clinical and economic value.
LIMITATIONS
There are several limitations to consider when interpreting the results of this study. First, identifying LOTs and tracking advancements of LOTs in patients with MM in the claims database poses challenges, and we have mitigated this by adapting the guideline outlined by Rajkumar et al. However, the identified LOTs are only approximation and could differ if a different guideline was used. Consequently, the use of different guidelines could potentially result in different study outcomes. Second, the MarketScan CCAE and MDCR databases are not representative of all patients in the United States, and the findings of this study may not be generalizable to patients with MM with insurance types not represented within the MarketScan database. Third, the inclusion of patients with variable follow-up periods can introduce bias in the study results. To mitigate this, we used the KMSA technique to adjust for varying follow-up durations. Fourth, the end of the study period was December 2019, and the study did not reflect the impact of newer therapies approved after the study period on the study outcomes. Future follow-up studies that assess the HCRU and costs incurred after exposure to CAR-T, ADC, or other newly approved therapies are warranted. Another limitation is that we did not formally test any hypotheses of differences in costs or utilization. Lastly, this study did not consider indirect costs, such as costs associated with absenteeism and presenteeism. Future research on indirect costs is warranted to fully characterize the economic burden associated with patients with MM who have been exposed to different drug classes and combinations.
Conclusions
This retrospective real-world study assessed and described HCRU and costs among patients with MM exposed to different drug classes and combinations. The study findings suggest that the economic burden following exposure to multiple drug classes and combinations is substantial, especially among the TCE patients and in ambulatory and inpatient settings. Total costs incurred by these patients were primarily MM-related and mainly attributable to MM drug and administration costs. These findings highlight the need for more effective treatments that can delay progression and mitigate the resource use and economic burden in the management of MM. With recent FDA approval for treatments with novel mechanisms, such as ADC, CAR-T therapies, and bispecific T-cell therapy, future research on the HCRU and costs incurred after exposure to these treatments is warranted.
REFERENCES
- 1.Bird, S, Boyd, K. Multiple myeloma: An overview of management. Palliat Care Soc Pract. 2019;13:1178224219868235. doi:10.1177/1178224219868235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.SEER. Cancer stat facts: Myeloma. National Cancer Institute. 2020. Accessed April 24, 2023. https://seer.cancer.gov/statfacts/html/mulmy.html [Google Scholar]
- 3.Nijhof IS, van de Donk NWCJ, Zweegman S, Lokhorst HM. Current and new therapeutic strategies for relapsed and refractory multiple myeloma: An update. Drugs. 2018;78(1):19-37. doi:10.1007/s40265-017-0841-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kumar SK, Callander NS, Adekola K, et al. Multiple Myeloma, Version 3.2021, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 18(12), 1685-717. doi:10.6004/jnccn.2020.0057 [DOI] [PubMed] [Google Scholar]
- 5.Wang F, Gorsh B, DerSarkissian M, et al. Treatment patterns and outcomes of patients with double-class refractory or triple-class refractory multiple myeloma: A retrospective US electronic health record database study. Leuk Lymphoma. 2023;64(2):398-406. doi:10.1080/10428194.2022.2140284 [DOI] [PubMed] [Google Scholar]
- 6.Yong K, Delforge M, Driessen C, et al. Multiple myeloma: Patient outcomes in real-world practice. Br J Haematol. 2016;175(2):252-64. doi:10.1111/bjh.14213 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Teitelbaum A, Ba-Mancini A, Huang H, et al. Health care costs and resource utilization, including patient burden, associated with novel-agent-based treatment versus other therapies for multiple myeloma: Findings using real-world claims data. Oncologist. 2013;18(1):37-45. doi:10.1634/theoncologist.2012-0113 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fonseca R, Abouzaid S, Bonafede M, et al. Trends in overall survival and costs of multiple myeloma, 2000-2014. Leukemia. 2017;31(9):1915-21. doi:10.1038/leu.2016.380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.MacEwan JP, Batt K, Yin W, et al. Economic burden of multiple myeloma among patients in successive lines of therapy in the United States. Leuk Lymphoma. 2018;59(4):941-9. doi:10.1080/10428194.2017.1361035 [DOI] [PubMed] [Google Scholar]
- 10.Hagiwara M, Panjabi S, Delea T, Yucel E, Fonseca R. Burden of disease progression in patients with multiple myeloma in the US. Leuk Lymphoma. 2020;61(1):47-55. doi:10.1080/10428194.2019.1648802 [DOI] [PubMed] [Google Scholar]
- 11.Hernández-Rivas JÁ, Ríos-Tamayo R, Encinas C, Alonso R, Lahuerta JJ. The changing landscape of relapsed and/or refractory multiple myeloma (MM): Fundamentals and controversies. Biomark Res. 2022;10(1):1. doi:10.1186/s40364-021-00344-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Madduri D, Hagiwara M, Parikh K, et al. Real-world treatment patterns, health care use and costs in triple-class exposed relapsed and refractory multiple myeloma patients in the USA. Future Oncol. 2021;17(5):503-15. doi:10.2217/fon-2020-1003 [DOI] [PubMed] [Google Scholar]
- 13.Jagannath S, Joseph N, He J, Crivera C, Fu AZ, Garret A, Shah N. Healthcare costs incurred by patients with multiple myeloma following triple class exposure (TCE) in the US. Oncol Ther. 2021;9(2):659-669. doi:10.1007/s40487-021-00175-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jagannath S, Joseph N, He J, et al. Healthcare costs of multiple myeloma patients with four or more prior lines of therapy, including triple-class exposure in the United States. Oncol Ther. 2022;10(2):411-20. doi:10.1007/s40487-022-00198-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Truven Health Analytics. IBM MarketScan Research Databases for life sciences researchers. IBM Watson Health. [White Paper]. https://www.ibm.com/downloads/cas/OWZWJ0QO
- 16.Rajkumar SV, Richardson P, San Miguel JF. Guidelines for determination of the number of prior lines of therapy in multiple myeloma. Blood. 2015;126(7):921-2 doi:10.1182/blood-2015-05-647636 [DOI] [PubMed] [Google Scholar]
- 17.US Bureau of Labor Statistics. Measuring price change in the CPI: Medical care. US Bureau of Labor Statistics. 2020. Accessed April 26, 2022. https://www.bls.gov/cpi/factsheets/medical-care.htm
- 18.Centers for Medicare & Medicaid Services. Place of service code set. Centers for Medicare & Medicaid Services. 2021. Accessed May 11, 2022. https://www.cms.gov/Medicare/Coding/place-of-service-codes/Place_of_Service_Code_Set
- 19.Chua KP, Fischer MA, Linder JA. Appropriateness of outpatient antibiotic prescribing among privately insured US patients: ICD-10-CM based cross sectional study. BMJ. 2019;364:k5092. doi:10.1136/bmj.k5092 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lin DY, Feuer EJ, Etzioni R, Wax Y. Estimating medical costs from incomplete follow-up data. Biometrics. 1997;53(2):419-34. [PubMed] [Google Scholar]
- 21.Wijeysundera HC, Wang X, Tomlinson G, Ko DT, Krahn MD. Techniques for estimating health care costs with censored data: An overview for the health services researcher. Clinicoecon Outcomes Res. 2012;4:145-55. doi:10.2147/CEOR.S31552 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wang PF, Yee CW, Gorsh B. et al. Treatment patterns and overall survival of patients with double-class and triple-class refractory multiple myeloma: A US electronic health record database study. Leuk Lymphoma. 2023;64(2):398-406. doi:10.1080/10428194.2022.2140284 [DOI] [PubMed] [Google Scholar]
- 23.Bruno AS, Willson JL, Opalinska JM, et al. Recent real-world treatment patterns and outcomes in US patients with relapsed/refractory multiple myeloma. Expert Rev Hematol. 2020;13(9):1017-25. doi:10.1080/17474086.2020.1800451 [DOI] [PubMed] [Google Scholar]
- 24.Hollmann S, Moldaver D, Goyert N, Grima D, Maiese EM. A U.S. Cost analysis of triplet regimens for patients with previously treated multiple myeloma. J Manag Care Spec Pharm. 2019;25(4):449-59. doi:10.18553/jmcp.2019.25.4.449 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Braunlin M, Belani R, Buchanan J, Wheeling T, Kim C. Trends in the multiple myeloma treatment landscape and survival: A U.S. analysis using 2011-2019 oncology clinic electronic health record data. Leuk Lymphoma. 2021;62(2):377-86. doi:10.1080/10428194.2020.1827253 [DOI] [PubMed] [Google Scholar]