Introduction
Multiple myeloma (MM) is a haematologic neoplasm characterized by the aberrant clonal proliferation of plasma cells. The age-standardized prevalence and incidence rates of MM worldwide are estimated to be approximately 6.88 and 1.60 per 100,000 persons, respectively [ 1 ]. In recent decades, the advent of novel agents, such as monoclonal antibodies (mAbs), proteasome inhibitors (PIs), and immunomodulatory drugs (IMiDs), has markedly extended the duration of progression-free survival (PFS) and overall survival (OS) among patients with MM, especially those under the age of 65 [ 2 – 4 ].
However, these advancements come at a high financial cost, contributing to financial toxicity (FT), which is increasingly recognized as a critical issue affecting patients’ well-being [ 5 – 7 ]. FT refers to the burden of out-of-pocket expenses, loss of income, and the financial distress associated with the cost of cancer treatment [ 8 ]. To assess this subjective financial burden, de Souza et al. developed and validated the Comprehensive Score for Financial Toxicity (COST) [ 9 ], a tool that measures the perceived financial distress resulting from treatment-related costs. Studies by Huntington et al. [ 10 ] and Banerjee R et al. [ 11 ] have shown that a considerable proportion of patients with MM in the U.S. experienced significant FT, regardless of whether they were receiving maintenance therapy or under observation.
MM remains a generally incurable disease. Patients with MM typically present pronounced symptom burdens, such as skeletal complications (including pain and fractures), recurrent bacterial infections, renal impairment, and anaemia [ 12 ]. Therefore, slowing disease progression and managing associated symptoms are key objectives in clinical practice to improve HRQoL. Additionally, FT has been recognized as a significant predictor of HRQoL. Accumulating evidence suggests that FT not only affects treatment adherence but also deteriorates HRQoL, leading to psychological distress, such as depression and anxiety, and ultimately reducing overall well-being [ 5 , 13 , 14 ]. Therefore, a systematic review by Stevanus Pangestu and Fanni Rencz revealed that substantial financial toxicity is associated with a decline in HRQoL [ 15 ]. This systematic review also identified a research gap, emphasizing the need for longitudinal investigations to elucidate the dynamic impact of financial toxicity on HRQoL.
Socioeconomic status (SES), which encompasses income, education, and occupation, plays a critical role in shaping patients’ financial resilience and access to supportive resources. Previous studies have shown that lower SES is associated with greater FT [ 16 – 18 ]. Therefore, the impact of FT on HRQoL may differ across SES groups. Patients with lower SES may be more vulnerable to FT because of limited financial reserves and restricted access to financial assistance programs. In contrast, those with higher SES may possess better financial coping mechanisms, which could buffer the adverse effects of FT on HRQoL [ 19 ]. Despite these theoretical considerations, the moderating role of SES in the relationship between FT and HRQoL remains underexplored among patients with MM.
To address these gaps, in this study, a longitudinal study was designed to investigate changes in FT and HRQoL over a 24-month follow-up period. We also assessed whether SES moderates the impact of FT on HRQoL. The findings of this study offer critical insights into the long-term consequences of financial burden in patients with MM and inform targeted interventions aimed at improving the financial well-being and quality of life among vulnerable patient populations.
Methods
Study design and participants
We conducted a longitudinal, multicentre, real-world study using data on HRQoL and financial toxicity among patients with MM in China. This study was approved by the Institutional Review Board of Zhejiang University School of Public Health (ZGL202008-1).
Participants completed the baseline assessment through face-to-face interviews at their initial clinic visit during the study period. Scheduled follow-up assessments were conducted at 3-, 6-, 12-, and 24-month visits. The initial assessments took place between March and May 2021, and the final follow-up was completed in July 2023. Owing to budgetary and logistical constraints, patients with MM were recruited from eight tertiary hospitals across four provinces in China: Zhejiang (4 hospitals), Xinjiang (2 hospitals), Anhui (1 hospital), and Guizhou (1 hospital). The main inclusion criteria were as follows: (1) a primary diagnosis of MM based on the 2014 International Myeloma Working Group (IMWG) criteria [ 20 ]; (2) age ≥ 18 years at the time of study enrolment; (3) voluntary participation with signed informed consent; and (4) the ability to communicate effectively with the investigator.
In this study, whether a participant received active therapy or was managed by observation was determined by the attending specialists on the basis of the individual’s clinical status. Patients were excluded if they had significant comorbidities (e.g., advanced Parkinson’s disease, poststroke conditions, malignant tumours, and end-stage renal disease), mental health disorders, or speech impairment or if they were pregnant or breastfeeding.
Measures
Demographics and clinical characteristics
At baseline, patient demographic data—such as age, sex, marital status, education level, insurance status, occupation, household registration, and annual family income—were collected via questionnaire. Clinical characteristics were extracted from electronic medical records, with the date of diagnosis, International Staging System (ISS) stage and ASCT history recorded at baseline and treatment line and disease status updated at each follow-up visit. The treatment line was determined using an algorithm from previous studies. The start of the first-line therapy was defined as the date of the first treatment with an anti-MM oral prescription or administration of anti-MM therapy. A new line of therapy is defined when a different regimen is initiated after the discontinuation of the previous line or if one or more drugs are unexpectedly added or substituted in the existing regimen. In our study, patients in “remission” refer to patients who achieve a complete response (CR) or a very good partial response (VGPR) defined by the IMWG [ 20 ]; generally, these patients remain on their current line of therapy.
HRQoL
HRQoL was assessed by the EuroQol-5 Dimensions-5 Levels (EQ-5D-5 L) and Functional Assessment of Cancer Therapy-Multiple Myeloma (FACT-MM) at each survey visit. The EQ-5D-5 L provides a standardized overall health utility perspective, evaluating dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, in addition to the visual analogue scale (VAS) score [ 21 , 22 ]. The EQ-5D-5 L health status can be converted into a single “utility” score using the Chinese scoring algorithm, which generates utility scores ranging from − 0.391 to 1.000 [ 23 ], with higher scores indicating better health status. The EQ-VAS score (0–100) aligns with the EQ-5D-5 L utility score, with higher values indicating better health status.
The FACT-MM is a disease-specific instrument that captures myeloma-related symptoms and functional concerns that are not covered by generic tools. FACT-MM combines the general version of the FACT (FACT-G) with an MM-specific subscale (MM domain) [ 24 , 25 ]. The FACT-G comprises subscales for physical well-being (7 items), social/family well-being (7 items), emotional well-being (6 items), and functional well-being (7 items). The MM domain includes 14 items and addresses symptom burden and disease-specific well-being. Summary scores can be calculated by adding the four subscale scores of the FACT-G and the MM domain. The total FACT-MM score ranges from 0 to 164, with higher scores indicating a better quality of life.
Financial toxicity
Financial toxicity was assessed using the COST measure at each survey visit. The COST is an 11-item patient-reported outcome measure developed for patients with advanced malignancies and provides a financial toxicity score ranging from 0 to 44, with higher scores indicating lower financial burden [ 9 ].
The use of the EQ-5D-5 L, FACT-MM, and COST measures was officially registered, and the validated Chinese versions of these instruments were employed in this study.
Assessment of SES
In accordance with previous studies, variables such as income, education and occupation were used to measure SES [ 26 – 29 ]. Given the substantial economic disparity between urban and rural areas in China, household registration was also included as an SES predictor. In this study, family income was categorized into quartiles (lowest, lower, higher, and highest). Education was classified as less than upper secondary, upper secondary, or tertiary, following the 2011 International Standard Classification of Education. Employment status was categorized as unemployed, employed, or retired, and household registration was categorized as either urban or rural.
To construct a composite index indicating socioeconomic status, we employed a latent class analysis (LCA) based on family annual income, education level, household registration and employment status. LCA is a widely used statistical method for identifying unobserved subgroups on the basis of multiple observed categorical variables [ 26 – 29 ]. In the LAC, the two-class model yielded the lowest AIC and BIC and achieved an entropy greater than 0.80, indicating good classification quality (Table S1 ). Therefore, two classes of SES (low and high) were employed in this study (Table S2).
Statistical analysis
Descriptive statistics were used to summarize patient characteristics. Frequencies and percentages are reported for categorical variables. The mean and standard deviation (SD) of HRQoL and COST scores were calculated for each patient at each follow-up visit. Additionally, univariate analyses were performed at baseline to examine associations between patient characteristics and HRQoL and COST; t tests were used for comparisons between two groups, and one-way ANOVA was used for comparisons among three or more groups.
Given the primary objective of this study, patients with missing data on any of the EQ-5D-5 L, FACT-MM, or COST items were excluded from the main analysis sample. Additionally, patients who were lost to follow-up, withdrew from the study, or died during the follow-up period were also excluded from the main analyses (Table S3).
Rösel et al. reported that linear mixed-effects models (LMMs) are well suited for longitudinal studies with missing data [ 30 ]. Guided by this evidence, we assumed that the missing-at-random assumption held in this longitudinal study. Accordingly, we applied LMMs with a random intercept for each individual to assess longitudinal changes in the COST and HRQoL scores. In the LMMs for COST and HRQoL, the key independent variables included follow-up visit, SES, treatment line, and disease status. Additionally, the HRQoL model incorporated COST scores and the interaction between SES and COST. Both models were adjusted for age, sex, marital status, insurance status, date of diagnosis, ISS stage, and ASCT history. Given that newly diagnosed patients with MM have typically not yet received systemic therapy and are therefore unlikely to have achieved remission, we treated the “newly diagnosed” category as equivalent to “first-line treatment without remission” in the LMMs. This assumption was made to reduce redundancy and to mitigate potential multicollinearity in the analysis. Owing to observed follow-up attrition, we performed sensitivity analyses to assess the robustness of our findings. These analyses applied multiple imputation (40 imputations) for missing COST, EQ-5D-5 L utility, and FACT-MM scores, excluding postdeath observations from imputation, and then reestimated the prespecified mixed-effects models.
All analyses were completed using R (version 4.2.2), with statistical significance assessed at the 0.05 level.
Results
Participant characteristics
Among the 516 eligible patients with MM who were recruited to participate in the study, 441 were enrolled and completed the baseline assessment. Owing to missing data, loss to follow-up, death and withdrawal, 412 (93.5%), 371 (84.1%), 335 (76.0%) and 253 (57.4%) patients were included in the main analytical datasets at the 3-, 6-, 12- and 24-month visits, respectively. The patient inclusion process is illustrated in Fig. 1 . Patient characteristics at baseline are presented in Table 1 . Among the 441 patients, 67.80% were < 65 years old, 55.10% were male, 91.61% were married, and 62.13% had less than a secondary school education. With respect to insurance, 58.05% held urban–rural resident basic medical insurance, and 39.46% held urban employee basic medical insurance. Most participants were retired (75.06%), had urban household registration (73.24%), were undergoing first-line treatment (45.12%), were in remission (68.03%) and did not undergo ASCT (71.88%).
Table 1
Patient characteristics of MM patients at baseline assessment
Note: UEBMI Urban Employee Basic Medical Insurance, URRBMI Urban-Rural Resident Basic Medical Insurance
| Characteristics | Patient ( n = 441) | |
| n | % | |
| Age | ||
| < 65 | 299 | 67.80% |
| ≥ 65 | 142 | 32.20% |
| Gender | ||
| Male | 243 | 55.10% |
| Female | 198 | 44.90% |
| Marital status | ||
| Single | 37 | 8.39% |
| Married | 404 | 91.61% |
| Education | ||
| Less than secondary school | 274 | 62.13% |
| Upper secondary | 109 | 24.72% |
| Tertiary | 58 | 13.15% |
| Insurance | ||
| URRBMI | 256 | 58.05% |
| UEBMI | 174 | 39.46% |
| Other | 11 | 2.49% |
| Occupational status | ||
| Employed | 74 | 16.78% |
| Unemployed | 36 | 8.16% |
| Retired | 331 | 75.06% |
| Household registration | ||
| Urban | 323 | 73.24% |
| Rural | 118 | 26.76% |
| Family annual income | ||
| Quartile 1(lowest) | 110 | 24.94% |
| Quartile 2 | 110 | 24.94% |
| Quartile 3 | 111 | 25.17% |
| Quartile 4 (highest) | 110 | 24.94% |
| Time since diagnosis | ||
| < 12 m | 176 | 39.91% |
| 12–36 m | 158 | 35.83% |
| 36–60 m | 76 | 17.23% |
| >60 m | 31 | 7.03% |
| Treatment lines | ||
| Newly diagnosed | 57 | 12.92% |
| First-line treatment | 198 | 45.12% |
| Second-line treatment | 144 | 22.68% |
| Third-line or later treatment | 42 | 19.27% |
| ISS stage | ||
| I | 189 | 42.86% |
| II | 147 | 33.33% |
| III | 95 | 21.54% |
| Unknown | 10 | 2.27% |
| Disease status | ||
| Remission | 300 | 68.03% |
| No remission | 141 | 31.97% |
| ASCT history | ||
| Yes | 124 | 28.12% |
| No | 317 | 71.88% |
HRQoL and financial toxicity
The distributions of the EQ-5D-5 L utility index, VAS score, FACT-MM, and COST score at baseline are shown in Fig. 2 . For the EQ-5D-5 L index, most respondents had utility values in the range of 0.80 to 0.85, with 99 patients reporting values in this range. A total of 59 patients reported utility values greater than 0.95, 48 of whom had a utility index of 1. For the EQ-VAS scores, the most frequently reported range was 85 to 90, with 98 respondents in this interval. The most frequently observed FACT-MM scores fell within the 120–130 range, with 93 patients reporting scores in this interval. Similarly, for COST scores, the highest number of respondents (151 patients) reported scores in the 20–25 range.
At baseline, the mean scores for the EQ-5D-5 L utility index, EQ-VAS, FACT-MM, and COST were 0.79 (SD = 0.25), 81.96 (SD = 18.73), 117.46 (SD = 21.76), and 21.84 (SD = 5.78), respectively (Table 2 ). In terms of HRQoL, better outcomes were observed among patients who were < 65 years, who were employed, who were diagnosed 12 to 36 months prior, who received first-line treatment, and who were classified as International Staging System (ISS) stage I, or who were in remission. With respect to financial toxicity, as measured by COST scores, higher scores (indicating lower financial toxicity) were reported by patients aged ≥ 65 years. Greater COST scores were also associated with being female, single, having tertiary-level education, being covered by the Urban Employee Basic Medical Insurance (UEBMI), being retired, residing in urban areas, being diagnosed within the past 12 months, being newly diagnosed, or being in remission. HRQoL and financial toxicity by treatment line and baseline disease status are summarized in Table S4.
Table 2
Univariate analyses of EQ-5D utility, EQ-VAS score, FACT-MM scores, and COST scores at baseline assessment
| Characteristics | EQ- 5D-5 L | EQ-VAS | FACT-MM | COST |
| Overall population | 0.79 (0.25) | 81.96 (17.73) | 117.46 (21.76) | 21.84 (5.78) |
| Age | ||||
| < 65 | 0.80 (0.14) | 83.24 (13.12) | 119.78 (17.84) | 19.90 (4.11) |
| ≥ 65 | 0.77 (0.15) | 79.26 (14.26) | 112.56 (19.23) | 25.92 (3.96) |
| P value | 0.040 | 0.004 | < 0.001 | < 0.001 |
| Gender | ||||
| Male | 0.80 (0.16) | 82.97 (12.34) | 118.36 (19.12) | 20.81 (3.89) |
| Female | 0.78 (0.19) | 80.71 (14.16) | 116.35 (20.79) | 23.11 (3.71) |
| P value | 0.231 | 0.074 | 0.292 | < 0.001 |
| Marital status | ||||
| Single | 0.77 (0.16) | 79.11 (16.61) | 116.21 (19.82) | 23.83 (4.11) |
| Married | 0.80 (0.19) | 82.69 (11.13) | 117.78 (17.32) | 21.33 (3.95) |
| P value | 0.169 | 0.015 | 0.457 | < 0.001 |
| Educational | ||||
| Less than secondary school | 0.77 (0.23) | 80.91 (17.12) | 116.31 (20.17) | 19.32 (4.91) |
| Upper secondary | 0.82 (0.19) | 82.56 (15.37) | 118.86 (18.09) | 25.49 (3.54) |
| Tertiary | 0.84 (0.16) | 85.78 (14.13) | 120.26 (17.21) | 26.89 (3.27) |
| P value | 0.198 | 0.227 | 0.299 | < 0.001 |
| Insurance | ||||
| UEBMI | 0.81 (0.19) | 83.18 (15.42) | 119.13 (16.12) | 22.61 (3.72) |
| URRBMI | 0.78 (0.21) | 80.03 (16.24) | 115.04 (20.21) | 20.72 (4.34) |
| P value | 0.124 | 0.042 | 0.020 | < 0.001 |
| Occupational status | ||||
| Employed | 0.83 (0.18) | 83.85 (13.23) | 121.26 (16.91) | 21.12 (4.01) |
| Unemployed | 0.77 (0.20) | 79.97 (16.03) | 115.84 (20.23) | 18.21 (3.62) |
| Retired | 0.78 (0.17) | 81.76 (15.12) | 116.78 (19.54) | 22.39 (3.89) |
| P value | 0.015 | < 0.001 | 0.036 | < 0.001 |
| Household registration | ||||
| Urban | 0.80 (0.20) | 82.43 (11.73) | 118.17 (17.51) | 22.46 (3.41) |
| Rural | 0.76 (0.19) | 80.68 (13.49) | 115.52 (19.12) | 20.13 (3.71) |
| P value | 0.019 | 0.236 | 0.169 | < 0.001 |
| Family annual income | ||||
| Quartile 1(lowest) | 0.75 (0.18) | 78.81 (15.54) | 114.13 (18.23) | 17.84 (3.89) |
| Quartile 2 | 0.77 (0.15) | 79.59 (13.82) | 117.05 (17.94) | 20.48 (3.91) |
| Quartile 3 | 0.82 (0.19) | 83.68 (13.79) | 118.79 (18.57) | 23.64 (3.83) |
| Quartile 4 (highest) | 0.83 (0.21) | 85.76 (14.48) | 119.87 (19.71) | 25.38 (4.07) |
| P value | 0.170 | 0.038 | 0.110 | < 0.001 |
| Time since diagnosis | ||||
| < 12 m | 0.79 (0.19) | 81.43 (14.06) | 115.46 (16.73) | 22.95 (5.09) |
| 12–36 m | 0.81 (0.18) | 83.67 (12.02) | 120.79 (17.23) | 21.84 (4.28) |
| 36–60 m | 0.78 (0.23) | 80.55 (13.61) | 116.52 (18.41) | 20.52 (4.65) |
| >60 m | 0.76 (0.21) | 79.67 (14.87) | 114.17 (17.53) | 18.76 (4.92) |
| P value | 0.013 | 0.022 | 0.012 | < 0.001 |
| Current treatment lines | ||||
| Newly diagnosed | 0.68 (0.21) | 72.34 (18.27) | 98. 51 (21.32) | 24.12 (4.95) |
| First-line treatment | 0.83 (0.15) | 84.54 (12.21) | 121.34 (18.21) | 23.36 (5.05) |
| Second-line treatment | 0.76 (0.18) | 80.83 (15.23) | 115.08 (17.21) | 21.13 (4.78) |
| Third-line or later treatment | 0.71 (0.16) | 75.53 (14.09) | 105.57 (19.54) | 18.11 (4.15) |
| P value | < 0.001 | < 0.001 | < 0.001 | < 0.001 |
| ISS stage | ||||
| I | 0.82 (0.15) | 84.89 (14.81) | 122.46 (16.27) | 21.76 (4.12) |
| II | 0.77 (0.19) | 80.38 (16.09) | 114.38 (17.15) | 22.13 (4.41) |
| III | 0.76 (0.17) | 78.58 (16.82) | 112.34 (16.93) | 21.54 (4.92) |
| P value | < 0.001 | < 0.001 | < 0.001 | 0.13 |
| Disease status | ||||
| Remission | 0.83 (0.13) | 86.57 (14.11) | 124.14 (15.72) | 23.54 (4.81) |
| No remission | 0.68 (0.17) | 72.14 (18.21) | 97.28 (21.96) | 18.23 (5.14) |
| P value | < 0.001 | < 0.001 | < 0.001 | < 0.001 |
| ASCT history | ||||
| Yes | 0.80 (0.19) | 82.53 (16.11) | 118.17 (20.83) | 22.47 (4.64) |
| No | 0.78 (0.21) | 81.74 (17.21) | 117.18 (19.01) | 21.60 (4.89) |
| P value | 0.14 | 0.29 | 0.21 | 0.09 |
During the follow-up period, a moderate increasing trend was observed in the EQ-5D-5 L utility values and FACT-MM scores, whereas a moderate decreasing trend was noted in the COST scores (Fig. 3 ). Furthermore, the results of the Mann–Whitney U test indicated significant differences in the HRQoL and COST scores between the baseline and follow-up assessments (Tables S5-S8). However, the HRQoL and COST scores at the 3-, 6-, and 12-month visits remained relatively stable, with no statistically significant differences observed among these time points. This finding indicates a stabilization in both quality of Life and financial toxicity between months 3 and 12, with no substantial improvement or deterioration.
The proportion of responses across each level of the EQ-5D-5 L dimensions is shown in Fig. 4 . Overall, improvements were observed across all five health dimensions over the course of follow-up. Among the five dimensions, anxiety/depression and pain/discomfort consistently exhibited the highest proportions of reported problems, whereas the self-care dimension consistently exhibited the lowest proportion of reported issues.
Mixed-effect models
In this study, the mixed-effects regression analysis included 1,812 longitudinal observations from 441 patients, and the results are presented in Tables 3 and 4 . Patients with high SES reported significantly higher COST scores than those with low SES did (β = 2.10, p < 0.001). Compared with baseline, COST scores significantly decreased at each follow-up visit, suggesting a worsening of financial toxicity over time: 3 months (β = − 1.38, p < 0.001), 6 months (β = − 1.55, p < 0.001), 12 months (β = − 2.08, p < 0.001), and 24 months (β = − 2.41, p < 0.001). Regarding treatment lines, patients who received second-line (β = −2.00, p < 0.001) and third-line or later treatment (β = −5.29, p < 0.001) had significantly lower COST scores than those who received first-line therapy did. Additionally, patients in remission demonstrated significantly higher COST scores than those who were not in remission did (β = 3.31, p < 0.001).
Table 3
Results of mixed-effects model regression for COST scores
Note: a measured at each follow-up visit; models were adjusted for age, gender, marital status, insurance, time since diagnosis, ISS stage and ASCT history
| Variable | β | SE | p |
| Constant | 19.80 | 2.02 | < 0.001 |
| SES | |||
| Low level | Reference | ||
| High level | 2.10 | 0.11 | < 0.001 |
| Follow-up point | |||
| Baseline | Reference | ||
| 3th month | −1.38 | 0.05 | < 0.001 |
| 6th month | −1.55 | 0.07 | < 0.001 |
| 12th month | −2.08 | 0.08 | < 0.001 |
| 24th month | −2.41 | 0.15 | < 0.001 |
| Treatment line a | |||
| First-line treatment | Reference | ||
| Second-line treatment | −2.00 | 0.07 | < 0.001 |
| Third-line or later treatment | −5.29 | 0.14 | < 0.001 |
| Disease status a | |||
| No remission | Reference | ||
| Remission | 3.31 | 0.16 | < 0.001 |
Table 4
Results of mixed-effects model regression for EQ-5D-5Lvalue and FACT-MM scores
Note: a measured at each follow-up visit; models were adjusted for age, gender, marital status, insurance, time since diagnosis, ISS stage and ASCT history
| Variable | EQ-5D-5 L | FACT-MM | ||||
| β | SE | p | β | SE | p | |
| Constant | 0.598 | 0.089 | < 0.001 | 81 | 6.12 | < 0.001 |
| COST | 0.006 | 0.0007 | < 0.001 | 1.60 | 0.10 | < 0.001 |
| SES | ||||||
| Low level | Reference | Reference | ||||
| High level | 0.130 | 0.003 | < 0.001 | 15.01 | 2.27 | < 0.001 |
| SES*COST | −0.001 | 0.0002 | < 0.001 | −0.12 | 0.003 | < 0.001 |
| Follow-up point | ||||||
| Baseline | Reference | Reference | ||||
| 3th month | 0.033 | 0.005 | 0.023 | 4.24 | 0.47 | 0.028 |
| 6th month | 0.048 | 0.007 | 0.009 | 6.13 | 0.43 | < 0.001 |
| 12th month | 0.054 | 0.011 | < 0.001 | 6.03 | 0.52 | < 0.001 |
| 24th month | 0.069 | 0.012 | < 0.001 | 8.87 | 0.61 | < 0.001 |
| Treatment line a | ||||||
| First-line treatment | Reference | < 0.001 | Reference | < 0.001 | ||
| Second-line treatment | −0.07 | 0.002 | < 0.001 | −5.94 | 0.15 | < 0.001 |
| Third-line or later treatment | −0.13 | 0.005 | < 0.011 | −15.26 | 2.89 | < 0.001 |
| Disease status a | ||||||
| No remission | Reference | Reference | ||||
| Remission | 0.171 | 0.023 | < 0.001 | 27.12 | 2.21 | < 0.001 |
Higher COST scores were also significantly associated with improved HRQoL, as reflected by both EQ-5D-5 L utility values (β = 0.006, p < 0.001) and FACT-MM scores (β = 1.60, p < 0.001). Patients with high SES demonstrated significantly higher EQ-5D-5 L utility values (β = 0.130; p < 0.001) and FACT-MM scores (β = 15.01; p < 0.001) than those with low SES did. The interaction between SES and COST scores was also significant for both outcomes (EQ-5D-5 L: β = −0.001, p < 0.001; FACT-MM: β = −0.12, p < 0.001), indicating that the positive effect of financial well-being on HRQoL was more pronounced among patients with lower SES. Compared with baseline, all follow-up visits were associated with significantly improved EQ-5D-5 L utility values and FACT-MM scores. In terms of treatment lines, patients who received second-line (EQ-5D-5 L: β = − 0.07, p < 0.001; FACT-MM: β = − 5.94, p < 0.001) and third-line or later therapies (EQ-5D-5 L: β = − 0.13, p = 0.011; FACT-MM: β = − 15.26, p < 0.001) reported significantly lower HRQoL than those who received first-line treatment did. Finally, patients in remission had significantly better HRQoL than those not in remission did (EQ-5D-5 L: β = 0.171, p < 0.001; FACT-MM: β = 27.12, p < 0.001). The sensitivity analysis results using MI were nearly identical to the primary analysis (Table S9-S10).
Discussion
In this study, we evaluated changes in HRQoL and financial toxicity among patients with MM in China over a two-year follow-up period. HRQoL was assessed using the EQ-5D-5 L, EQ-VAS, and FACT-MM, while financial toxicity was measured with the COST instrument at each follow-up point. To our knowledge, this is the first real-world study to longitudinally assess both HRQoL and financial toxicity among Chinese patients with MM and to examine the independent effects of follow-up visits, treatment lines, disease status, and SES on these outcomes. These findings provide important evidence to inform strategies for improving long-term patient-centred care and contribute to a better understanding of the economic value of clinical treatments in health economic evaluations.
During the 24-month follow-up period, HRQoL of patients with MM moderately improved. This pattern aligns with the typical clinical trajectory of HRQoL in MM, which can be broadly divided into three distinct phases [ 31 , 32 ]. HRQoL is markedly impaired at diagnosis because of bone involvement and fractures, recurrent infections, renal dysfunction, anaemia, and mood disorders [ 12 ]. Following the initiation of treatment, effective disease control and symptom relief could lead to significant improvements in HRQoL. Given that first- and second-line therapies—particularly during the maintenance phase—can induce deep and durable responses (with median PFS reaching up to 60 months in first-line and 24 months in second-line settings), the associated improvements in HRQoL are likely to be sustained over an extended period [ 33 ]. In the later stages of the disease, particularly among patients with relapsed or refractory MM, HRQoL tends to deteriorate again, mirroring the decline typically observed in patients with advanced-stage cancers [ 34 ]. In our study, 57 (12.92%) newly diagnosed patients and 42 (19.27%) patients who underwent third-line or later treatments were included. At baseline, HRQoL was lowest, primarily because of the impaired health status of newly diagnosed patients. By the 3-month follow-up, modest but clinically meaningful improvements in HRQoL were observed compared with baseline. From 3 to 12 months, HRQoL scores remained relatively stable. At the 24-month follow-up visit, patients had the highest HRQoL scores. This may be attributed to the fact that the majority of participants were enrolled during the early stages of the disease and remained in remission throughout the two-year follow-up period.
Abonour R and colleagues reported that patients with MM receiving maintenance therapy had a utility value of 0.853 [ 35 ], which is close to that observed in patients in remission receiving first-line treatment in our study (0.84). Another study by Hatswell et al. reported utility values of 0.530 for newly diagnosed patients, 0.646 for patients receiving first-line treatments, 0.591 for patients receiving second-line treatments, 0.578 for patients receiving third-line treatments, and 0.469 for patients receiving fourth-line or later treatments [ 36 ]. The relatively low utility values reported in Hatswell’s study may be explained by a relatively large proportion of patients in the progressive disease phase. A study conducted in 2020 by Li et al. demonstrated that EQ-5D-3 L utility values for Chinese patients with MM < 65 and ≥ 65 were 0.68 and 0.67, respectively, which were lower than those in this study [ 37 ]. The higher utility values in our study may also be explained by the introduction of novel anti-MM therapies—particularly daratumumab—and the use of denosumab for osteoporosis management, both of which have improved quality of life among patients with MM in recent years [ 38 ].
During the long-term treatment of MM, patients often develop considerable psychological stress because of the fear of disease relapse. Additionally, the main symptoms of MM include chest pain and lower back pain, which tend to intensify after physical activity. These clinical characteristics are consistent with our findings, in which anxiety/depression was the most frequently reported problematic domain in the EQ-5D-5 L, followed by pain/discomfort. These observations align with those of previous studies, which reported elevated psychological distress across the disease continuum [ 39 , 40 ]. Given these findings, we recommend that psychological support and pain management be integrated into routine care for patients with MM, which may collectively improve patients’ overall HRQoL.
In this study, despite improvements in HRQoL, financial toxicity significantly increased over time. Consistent with our findings, a study by Scott F. Huntington et al. [ 10 ], which included patients receiving at least three months of antimyeloma treatment at a tertiary academic centre in the USA, reported a higher average COST score of 23, indicating lower financial toxicity. Similarly, a study by Rahul Banerjee et al. [ 11 ] reported a higher average COST score of 31, likely reflecting a lower financial burden, as 84% of the participants were undergoing maintenance therapy or active surveillance. In China, the relatively recent market entry of novel antimyeloma agents—such as ixazomib (approved in 2018), daratumumab (2019) and carfilzomib (2021)—has resulted in high drug prices, worsening the financial burden on patients. Moreover, to maintain disease control, patients often require prolonged treatment with these expensive agents, which may lead to cumulative financial toxicity, as observed in our study.
The mixed-effects model revealed a strong positive association between financial well-being and HRQoL, indicating that patients with better quality of life tended to have lower financial toxicity. SES also emerged as a key determinant of both financial toxicity and HRQoL. Specifically, the interaction effect showed that improvements in financial well-being (higher COST scores) were associated with greater gains in HRQoL among patients with lower SES, suggesting that financially vulnerable patients may be more sensitive to changes in financial burden. These findings highlight the potential value of interventions targeting financial toxicity, particularly for socioeconomically disadvantaged groups. Our results are consistent with those of previous studies showing that lower SES is associated with greater financial toxicity among adult women with breast cancer [ 41 ] and that individuals with stroke who have lower SES have poorer overall HRQoL [ 42 ].
Disease status and treatment lines were also key determinants of both HRQoL and financial toxicity. Patients in remission reported significantly better HRQoL and lower levels of financial toxicity than those without remission did. This likely reflects not only a reduced symptom burden but also decreased treatment intensity during remission phases. In addition, patients who underwent second-line or third-/late-line therapy experienced markedly greater financial toxicity and poorer quality of life. These findings reflect the increasing complexity, intensity, and cost of treatment regimens in later lines, which often include combination therapies, novel agents, or prolonged hospitalization. Recent clinical trials have shown progressively decreasing clinical outcomes—measured by survival and QALYs—across successive lines of therapy [ 43 , 44 ], while health economic evaluations have revealed a corresponding stepwise increase in treatment costs and health care resource utilization [ 45 – 47 ]. These findings highlight the need to integrate financial counselling and support services into routine care for patients with MM, particularly those receiving later-line therapies.
Despite the contributions of this study to the understanding of HRQoL and financial toxicity among patients with MM, several limitations should be acknowledged. First, we assumed that missing data—arising from loss to follow-up, withdrawal, or other reasons—followed a missing-at-random mechanism. However, this assumption may not fully hold in real-world settings. In particular, patients with more severe disease may forgo active antimyeloma treatment or seek care at higher-level medical centres, resulting in loss to follow-up. Therefore, patients excluded from the analysis were more likely to have lower SES and poorer quality of life, potentially leading to an underestimation of financial toxicity and HRQoL impairments. Nevertheless, we adopted linear LMMs under the MAR assumption because baseline covariates were virtually complete, and LMMs exploit all available longitudinal observations without restricting the analysis to complete cases.
Second, participant recruitment was conducted exclusively in outpatient clinics, thereby excluding hospitalized patients. Given that inpatients often experience more severe disease symptoms, greater HRQoL impairments, and greater financial burdens, the omission of this subgroup may have led to an underestimation of both HRQoL impairments and financial toxicity in the overall MM population. Consequently, this selection bias could limit the generalizability of our findings to the broader MM population, especially those requiring hospitalization.
Third, given the relatively long survival of patients with MM, the two-year follow-up period in this study is relatively short and does not fully capture quality of life and financial toxicity at the advanced disease stage. Future research should extend follow-up durations to capture the long-term progression of HRQoL and financial toxicity in patients with MM while prioritizing evaluations among hospitalized individuals, especially those undergoing ASCT or salvage chemotherapy.
Conclusion
This study provides valuable evidence regarding HRQoL and financial toxicity as experienced by patients with MM in China. The results indicated a gradual improvement in HRQoL over time, whereas financial toxicity worsened. Among the HRQoL dimensions, anxiety/depression and pain/discomfort were the most commonly affected, highlighting the need to integrate psychosocial support and pain management into routine MM care. Notably, financial toxicity had a more pronounced negative effect on HRQoL among patients with lower SES. Therefore, financial counselling and support services should be prioritized for patients from lower SES backgrounds and those undergoing later-line treatments.
Supplementary Information
Supplementary Material 1.
Acknowledgements
The authors would like to thank the investigational staff and patients who participated in this study.
Abbreviations
Autologous Stem Cell Transplantation
COmprehensive Score for financial Toxicity
EuroQol-5 Dimensions-5 Level
Functional Assessment of Cancer Therapy-General
Functional Assessment of Cancer Therapy-Multiple Myeloma
Health-Related Quality of Life
Immunomodulatory Drugs
Monoclonal Antibodies
Multiple Myeloma
Maintenance Therapy
Overall Survival
Progression-Free Survival
Proteasome Inhibitors
Standard Deviation
Alongside the Visual Analogue Scale
Socioeconomic status
Authors’ contributions
AS and HD designed and supervised the study. WW, YW, and YW performed the data analysis. WW, WY, FT, MZ and QW contributed to the interpretation of data. WW and YW wrote the first draft of the manuscript and WW, YW and AS revised it critically. All authors read and approved the final manuscript.
Funding
This work was supported by the National Natural Science Foundation of China (52273308).
Data availability
The data analysed during current study are available from the corresponding author on reasonable request.
Declarations
Ethics approval nd consent to participate
This study was approved by the institutional review board of Zhejiang University School of Public Health and performed according to the Declaration of Helsinki. Ethics approval number: ZGL202008-1. Written informed consent was obtained for all participants in cross-section study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Weijia Wu and Yannan Wang contributed equally to this work.
Change history
10/11/2025
The original article has been updated to correct an erroneous affiliation.
Contributor Information
Hengjin Dong, Email: donghj@zju.edu.cn.
Aizong Shen, Email: anhuiaizongs@163.com.
References
1. Cowan AJ, Allen C, Barac A, et al. Global burden of multiple myeloma: A systematic analysis for the global burden of disease study 2016. JAMA Oncol. 2018;4(9):1221–7.
2. Braggio E, Kortüm KM, Stewart AK. SnapShot: multiple myeloma. Cancer Cell. 2015;28(5):678–e6781.
3. Turesson I, Velez R, Kristinsson SY, Landgren O. Patterns of improved survival in patients with multiple myeloma in the twenty-first century: a population-based study. J Clin Oncol. 2010;28(5):830–4.
4. Kumar SK, Dispenzieri A, Lacy MQ, et al. Continued improvement in survival in multiple myeloma: changes in early mortality and outcomes in older patients. Leukemia. 2014;28(5):1122–8.
5. Jiang H, Lyu J, Mou W, Jiang Q, Du J. Association between financial toxicity and health-related quality of life in cancer survivors: A systematic review. Asia Pac J Clin Oncol. 2023;19(4):439–57. 10.1111/ajco.13901.
6. Ver Hoeve ES, Ali-Akbarian L, Price SN, Lothfi NM, Hamann HA. Patient-reported financial toxicity, quality of life, and health behaviors in insured US cancer survivors. Support Care Cancer. 2021;29(1):349–58. 10.1007/s00520-020-05468-z.
7. Ghazal LV, Abrahamse P, Ward KC, Morris AM, Hawley ST, Veenstra CM. Financial toxicity and its association with Health-Related quality of life among partners of colorectal cancer survivors. JAMA Netw Open. 2023;6(4):e235897. 10.1001/jamanetworkopen.2023.5897. Published 2023 Apr 3.
8. Chan A, Ke Y, Tanay M, et al. Financial toxicity in cancer supportive care: an international survey. JCO Glob Oncol. 2024;10:e2400043. 10.1200/GO.24.00043.
9. de Souza JA, Yap BJ, Wroblewski K, Blinder V, Araújo FS, Hlubocky FJ, Nicholas LH, O’Connor JM, Brockstein B, Ratain MJ, et al. Measuring financial toxicity as a clinically relevant patient-reported outcome: the validation of the comprehensive score for financial toxicity (COST). Cancer. 2017;123:476–84.
10. Huntington SF, Weiss BM, Vogl DT, Cohen AD, Garfall AL, Mangan PA, Doshi JA, Stadtmauer EA. Financial toxicity in insured patients with multiple myeloma: a cross-sectional pilot study. Lancet Haematol. 2015;2(10):e408–16.
11. Banerjee R, Cowan AJ, Ortega M, Missimer C, Carpenter PA, Oshima MU, Salit RB, Vo PT, Lee CJ, Mehta RS, Kuderer NM, Shankaran V, Lee SJ, Su CT. Financial toxicity, time toxicity, and quality of life in multiple myeloma. Clin Lymphoma Myeloma Leuk. 2024;24(7):446–e4543.
12. Ntanasis-Stathopoulos I, Malandrakis P, Gavriatopoulou M. Practice-changing updates on multiple myeloma: highlights from the 2024 ASH annual meeting. J Hematol Oncol. 2025;18(1):54. 10.1186/s13045-025-01711-4. Published 2025 May 13.
13. Dar MA, Chauhan R, Trivedi V, Kumar R, Dhingra S. Assessing the prevalence of financial toxicity, its predictors and association with Health- related quality of life among radiation oncology patients in india: A Cross-Sectional patient reported outcome study. Int J Radiat Oncol Biol Phys. 2023;116(1):157–65. 10.1016/j.ijrobp.2022.11.040.
14. Hsu ML, Boulanger MC, Olson S, et al. Unmet needs, quality of life, and financial toxicity among survivors of lung cancer. JAMA Netw Open. 2024;7(4):e246872. 10.1001/jamanetworkopen.2024.6872. Published 2024 Apr 1.
15. Pangestu S, Rencz F. Comprehensive score for financial toxicity and Health-Related quality of life in patients with cancer and survivors: A systematic review and Meta-Analysis. Value Health. 2023;26(2):300–16. 10.1016/j.jval.2022.07.017.
16. Nogueira LA, Lenhani BE, Tomim DH, Kalinke LP. Financial toxicity. Asian Pac J Cancer Prev. 2020;21(2):289–93. 10.31557/APJCP.2020.21.2.289. Published 2020 Feb 1.
17. Mols F, Tomalin B, Pearce A, et al. Financial toxicity and employment status in cancer survivors. A systematic literature review. Support Care Cancer. 2020;28:5693–708.
18. Liu M, Huang D, Liu Y. Financial toxicity of patients with lung cancer in china: results from a National survey study. Cancer Med. 2023;12(4):4751–60. 10.1002/cam4.5244.
19. Marshall IJ, Wang Y, Crichton S, McKevitt C, Rudd AG, Wolfe CD. The effects of socioeconomic status on stroke risk and outcomes. Lancet Neurol. 2015;14:1206–18.
20. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International myeloma working group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15(12):e538–48. 10.1016/S1470-2045(14)70442-5.
21. Herdman M, Gudex C, Lloyd A, et al. Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res. 2011;20(10):1727–36. 10.1007/s11136-011-9903-x.
22. EuroQol Research Foundation. EQ-5D-5L User Guide, Version 3.0. 2019. https://euroqol.org/publications/user-guides . Accessed 16 Aug 2022.
23. Luo N, Liu G, Li M, et al. Estimating an EQ-5D-5L value set for China[J]. Value Health. 2017;20(4):662–69.
24. Weiss M, Jacobus S, Wagner LI, Cella D, Katz MS, Rajkumar V, et al. Development of the functional assessment of cancer therapy-multiple myeloma (FACT-MM) scale and validation in the Eastern cooperative oncology group trial E1A05. Blood. 2011;118(21):4184.
25. Webster K, Cella D, Yost K. The functional assessment of chronic illness therapy (FACIT) measurement system: properties, applications, and interpretation. Health Qual Life Outcomes. 2003;1(1):79.
26. Zhang YB, Chen C, Pan XF, et al. Associations of healthy lifestyle and socioeconomic status with mortality and incident cardiovascular disease: two prospective cohort studies. BMJ. 2021;373:604. 10.1136/bmj.n604. Published 2021 Apr 14.
27. Wang K, Fang Y, Zheng R, et al. Associations of socioeconomic status and healthy lifestyle with incident dementia and cognitive decline: two prospective cohort studies. EClinicalMedicine. 2024;76:102831. 10.1016/j.eclinm.2024.102831. Published 2024 Sep 16.
28. Wang Y, Liu M, Yang F, Chen H, Wang Y, Liu J. The associations of socioeconomic status, social activities, and loneliness with depressive symptoms in adults aged 50 years and older across 24 countries: findings from five prospective cohort studies. Lancet Healthy Longev. 2024;5(9):100618. 10.1016/j.lanhl.2024.07.001.
29. Li R, Li R, Xie J, et al. Associations of socioeconomic status and healthy lifestyle with incident early-onset and late-onset dementia: a prospective cohort study. Lancet Healthy Longev. 2023;4(12):e693–702. 10.1016/S2666-7568(23)00211-8.
30. Rösel I, Serna-Higuita LM, Al Sayah F, et al. What difference does multiple imputation make in longitudinal modeling of EQ-5D-5L data? Empirical analyses of simulated and observed missing data patterns. Qual Life Res. 2022;31(5):1521–32. 10.1007/s11136-021-03037-3.
31. Martino M, Rossi M, Ferreri A, et al. Quality of life outcomes in multiple myeloma patients: a summary of recent clinical trials. Expert Rev Hematol. 2019;12(8):665–84. 10.1080/17474086.2019.1634541.
32. Chakraborty R, Hamilton BK, Hashmi SK, Kumar SK, Majhail NS. Health-Related quality of life after autologous stem cell transplantation for multiple myeloma. Biol Blood Marrow Transpl. 2018;24(8):1546–53. 10.1016/j.bbmt.2018.03.027.
33. Silbermann R, Laubach J, Kaufman JL, et al. Health-related quality of life in transplant-eligible patients with newly diagnosed multiple myeloma treated with daratumumab, lenalidomide, bortezomib, and dexamethasone: Patient-reported outcomes from GRIFFIN. Am J Hematol. 2024;99(7):1257–68. 10.1002/ajh.27326.
34. Plesner T, Dimopoulos MA, Oriol A, et al. Health-related quality of life in patients with relapsed or refractory multiple myeloma: treatment with daratumumab, lenalidomide, and dexamethasone in the phase 3 POLLUX trial. Br J Haematol. 2021;194(1):132–9. 10.1111/bjh.17435.
35. Abonour R, Wagner L, Durie BGM, Jagannath S, Narang M, Terebelo HR, et al. Impact of post-transplantation maintenance therapy on health-related quality of life in patients with multiple myeloma: data from the connect ® MM registry. Ann Hematol. 2018;97(12):2425–36.
36. Hatswell AJ, Burns D, Baio G, et al. Frequentist and bayesian meta-regression of health state utilities for multiple myeloma incorporating systematic review and analysis of individual patient data. Health Econ. 2019;28:653–65.
37. Li X, Liu J, Chen M, et al. Health-related quality of life of patients with multiple myeloma: A real-world study in China. Cancer Med. 2020;9(21):7896–913. 10.1002/cam4.3391.
38. Kendler DL, Cosman F, Stad RK, Ferrari S. Denosumab in the treatment of osteoporosis: 10 years later: A narrative review. Adv Ther. 2022;39(1):58–74.
39. O’Donnell EK, Shapiro YN, Yee AJ, et al. Quality of life, psychological distress, and prognostic perceptions in patients with multiple myeloma. Cancer. 2022;128(10):1996–2004. 10.1002/cncr.34134.
40. Politi MC, Yen RW, Elwyn G, et al. Women who are young, Non-White, and with lower socioeconomic status report higher financial toxicity up to 1 year after breast cancer surgery: A Mixed-Effects regression analysis. Oncologist. 2021;26(1):e142–52. 10.1002/onco.13544.
41. Sun YA, Kalpakavadi S, Prior S, et al. Socioeconomic status and health-related quality of life after stroke: a systematic review and meta-analysis. Health Qual Life Outcomes. 2023;21(1):115. 10.1186/s12955-023-02194-y. Published 2023 Oct 25.
42. Yong K, Delforge M, Driessen C, et al. Multiple myeloma: patient outcomes in real-world practice. Br J Haematol. 2016;175(2):252–64. 10.1111/bjh.14213.
43. Delea TE, Ma Q, Kroog GS, et al. From Triple- to Penta-Exposed multiple myeloma: A Real-World study in a medicare population. Oncol Ther. 2024;12(3):565–83. 10.1007/s40487-024-00291-6.
44. Fu S, Wu CF, Wang M, Lairson DR. Cost effectiveness of transplant, conventional chemotherapy, and novel agents in multiple myeloma: A systematic review. PharmacoEconomics. 2019;37(12):1421–49.
45. Bayani DB, Lin YC, Nagarajan C, et al. Modeling First-Line daratumumab use for newly diagnosed, Transplant-Ineligible, multiple myeloma: A Cost-Effectiveness and risk analysis for healthcare payers. Pharmacoecon Open. 2024;8(5):651–64.
46. Blommestein HM, Franken MG, van Beurden-Tan CHY, et al. Cost-effectiveness of novel treatment sequences for Transplant-Ineligible patients with multiple myeloma. JAMA Netw Open. 2021;4(3):e213497.
47. Odejide OO, Cronin AM, Gray TF, et al. Anxiety and depression among patients newly diagnosed with lymphoma and myeloma. Blood Adv. 2025;9(7):1618–29. 10.1182/bloodadvances.2024014821.
Associated Data
Supplementary Materials
Supplementary Material 1.
Data Availability Statement
The data analysed during current study are available from the corresponding author on reasonable request.