Treatment Patterns and Clinical Outcomes Among Chinese Patients With Newly Diagnosed Multiple Myeloma: A Multicenter Retrospective Study (CHAMP‐MM)

ABSTRACT

The therapeutic landscape for newly diagnosed multiple myeloma (NDMM) in China has evolved significantly with wider access to novel agents, including proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), and CD38 monoclonal antibodies. However, real‐world data on treatment patterns and outcomes, particularly among high‐risk subgroups, remain limited. The CHAMP‐MM study aimed to evaluate real‐world frontline and post‐initial relapse treatment patterns and outcomes in NDMM patients, including high‐risk subgroups, defined as those age ≥ 65 (elderly), with renal impairment (RI), and with 1q cytogenetic abnormalities. The study utilized electronic medical record data from five tertiary hematologic centers in China. Adult patients diagnosed with active MM between 2018 and 2022 were included and followed until last visit or end of study (December 31, 2023). A total of 1139 patients were analyzed, including 364 elderly, 148 with RI, and 334 with 1q abnormalities. PI+IMiD‐based regimens were the most common frontline therapy (50.0%) overall, as well as in elderly patients and 1q subgroups, whereas PI‐based regimens were more frequently used in RI patients. Only 25.1% of NDMM patients received autologous stem cell transplant after induction. Following initial relapse, PI+IMiD‐based (32.6%) and CD38‐based (30.9%) regimens predominated. Frontline overall response rates were approximately 80% across all groups, whereas the proportion of patients achieving very good partial response (PR) or better ranged from 60.9% to 75.1% across subgroups. While overall 12‐ and 24‐month progression‐free survival (PFS) rates remained high, outcomes were poorer in high‐risk subgroups. This study reflects real‐world treatment trends in China over 5 years and emphasizes the continued need for tailored treatment approaches in high‐risk populations.

1. Introduction

Multiple myeloma (MM) is the second most common hematologic malignancy, primarily affecting the elderly population. In urban areas of China, it has a national prevalence of 6.88 per 100,000 and an incidence of 1.60 per 100,000 person‐years [1]. Over recent decades, the widespread clinical use of proteasome inhibitors (PIs), immunomodulatory drugs (IMiDs), CD38 monoclonal antibodies, and autologous stem cell transplantation (ASCT) has markedly improved patient prognosis [2, 3]. However, access to advanced therapies remains limited in many regions, and the heterogeneity of MM in terms of patient characteristics and risk stratification continues to pose challenges, particularly in developing countries such as China [4, 5]. In China, novel agents including bortezomib, lenalidomide, and daratumumab have been included in the national medical insurance system since 2016, significantly improving clinical accessibility. Given the expanded access to novel agents in recent years, it is likely that treatment patterns have evolved substantially. However, to date, no comprehensive studies have been conducted to assess the current real‐world treatment landscape in China, and data are lacking on how these changes in treatment patterns have impacted patient outcomes.

Although survival rates have continued to improve with advances in treatment, age at diagnosis remains a key determinant of prognosis [6]. In current Chinese clinical practice, patients aged ≥ 65 years are generally considered ineligible for ASCT. However, an optimal treatment strategy for elderly MM patients has yet to be established, and few real‐world studies have focused on this subpopulation in China [7]. Renal impairment (RI) is one of the most common complications and clinical manifestations of MM, affecting approximately 25%–50% of patients at some point during the disease course [8]. The presence of RI can influence physicians' decisions regarding the initial use of lenalidomide and the feasibility of ASCT, potentially impacting induction treatment outcomes. Nonetheless, large‐scale data supporting these associations are currently lacking [9]. Genetic abnormalities in MM cells are critical drivers of disease progression. Among these, gain or amplification of 1q is one of the most frequent cytogenetic abnormalities, occurring in approximately 40% of newly diagnosed multiple myeloma (NDMM) cases [10, 11]. Although 1q abnormalities have been shown to correlate with early disease progression and resistance to anti‐myeloma therapy, data from China on this high‐risk subpopulation remain limited due to a lack of published research [12].

Despite the widespread adoption of novel agents, limited information is available regarding the clinical characteristics and treatment outcomes of MM patients in China, particularly those with high‐risk features such as advanced age, renal impairment, and 1q abnormalities. To address this gap, the present study aimed to generate real‐world evidence on treatment patterns and clinical outcomes among the overall MM population and key high‐risk subgroups through a multicenter, retrospective cohort study.

2. Materials and Methods

2.1. Study Design and Patients

The CHAMP‐MM study was conducted using the electronic medical record (EMR) data from five tertiary hematology centers across China. Patients aged ≥ 18 years who were initially diagnosed with active MM between 2018 and 2022, and had continuous healthcare encounters at study sites were included. Patients who were diagnosed with smoldering multiple myeloma (SMM), missed initial diagnosis information, treated at other hospitals, or had missing electronic medical records due to hospital information system (HIS) transitions were excluded. Baseline was defined as 14 days prior to the initial diagnosis of MM. Follow up was defined as the period from the initial diagnosis of MM to last visit or end of study (December 31, 2023).

The elderly subgroup was defined as patients aged ≥ 65 years at baseline. Patients with an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m², at baseline were defined as the RI subgroup. Patients exhibiting gain or amplification of 1q, as identified by fluorescence in situ hybridization (FISH) at diagnosis, were defined as the 1q abnormalities subgroup. Cytogenetic data from FISH testing at diagnosis were available for a subset of patients; accordingly, all analyses involving 1q abnormalities were restricted to patients with available FISH results. Additional subgroup analyses by sex were performed to explore potential differences in treatment selection. The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of all participating hospitals. Informed consent was waived due to the retrospective nature of the study.

2.2. Study Variables and Outcomes

Baseline characteristics collected included age, sex, M‐protein isotype, International Staging System stage (ISS). Frontline and post first relapse induction regimens were categorized as CD38 antibody‐, PI+IMiD‐, PI‐, IMiD‐based, or conventional. CD38‐based regimens referred to daratumumab‐containing therapies, either as monotherapy or in combination with PIs and/or IMiDs. PI+IMiD‐based regimens included combinations of a PI (bortezomib, ixazomib, or carfilzomib) with an IMiD (lenalidomide, pomalidomide, or thalidomide) plus corticosteroids, with or without conventional cytotoxic agents (e.g., cyclophosphamide, doxorubicin, or melphalan). PI‐based regimens consisted of PI‐containing doublet or triplet combinations without IMiDs, administered with corticosteroids, with or without conventional cytotoxic agents. IMiD‐based regimens comprised IMiD‐containing combinations with corticosteroids, with or without conventional cytotoxic agents. Conventional regimens were defined as treatments not containing CD38 monoclonal antibodies, PIs, or IMiDs, including melphalan‐, cyclophosphamide‐, or doxorubicin‐based combinations with corticosteroids. Transplantation status was recorded. A new line of therapy was defined by: (1) initiation of a different regimen after discontinuation of all prior agents; (2) unplanned addition or substitution of ≥ 1 agent in the regimen (excluding initial modifications occurring within 30 days of first agent); or (3) underwent > 1 stem cell transplantation (SCT), unless part of a planned tandem SCT within 90 days [13].

Response outcomes were assessed by physicians according to the International Myeloma Working Group uniform response criteria [14]. Overall response rate (ORR) included partial response (PR) or better; ≥ VGPR rate included very good PR, complete response (CR), and stringent CR (sCR). Response assessment was based on the optimal response achieved within a treatment line, defined as the best overall response (BOR). Progression‐free survival (PFS) was defined as time from anti‐myeloma therapy initiation to first documented disease progression, death from any cause, or last follow‐up.

2.3. Statistical Methods

Descriptive analyses of patient demographic and clinical characteristics, and treatment patterns were conducted. Differences in treatment characteristics between groups were assessed using the chi‐squared test. The number and percentage of patients achieving ORR and VGPR or better, as well as the response in each category, were summarized. Median PFS with its 95% confidence interval (CI) was analyzed using the Kaplan–Meier method. PFS rates at 12 months and 24 months were also provided with 95% CIs. All patients without an event (disease progression or death) were censored at the earlier of the start of the next line of therapy or the date of last visit before the end of study. The proportional hazards assumption was assessed using Schoenfeld residuals derived from Cox proportional hazards models. Exploratory univariable and multivariable Cox proportional hazards regression analyses were additionally performed to evaluate baseline factors associated with progression‐free survival. In the overall cohort, the multivariable model included age, sex, ISS stage, and baseline renal impairment. A corresponding analysis restricted to patients with available baseline FISH data additionally included 1q abnormality as a binary variable (positive vs. negative) to evaluate its association with PFS after adjustment for these baseline clinical factors. A separate exploratory analysis was conducted to examine heterogeneity in prognosis according to the cytogenetic context of 1q abnormalities. Patients with available baseline FISH results were categorized as without 1q abnormalities, isolated 1q abnormalities, or 1q abnormalities co‐occurring with additional high‐risk cytogenetic abnormalities (HRCAs), including del17p, t (4;14), and t (14;16). In this analysis, the three‐category cytogenetic classification was the exposure of primary interest, and the multivariable Cox model was adjusted for age, sex, ISS stage, baseline CCI, and M‐protein subtype to minimize the loss of analyzable patients associated with missing baseline RI data. A sensitivity analysis was subsequently performed using age, sex, ISS stage, and baseline RI as covariates while retaining the same three‐category cytogenetic classification. All statistical analyses were performed using R version 4.3.3 (R Foundation for Statistical Computing, Vienna, Austria).

3. Results

3.1. Patient Characteristics

A total of 3608 MM patients were identified between January 1, 2018, and December 31, 2022, of whom 1139 met the eligibility criteria for inclusion (Figure 1). Among them, 32.0% (364/1139) were aged ≥ 65 years. Renal impairment at baseline was observed in 26.2% (148/566) of patients. FISH data were available in 673 patients (59.1% of the total cohort), among whom 334 (49.6%) had 1q abnormalities. Baseline characteristics of the overall population and subgroups are summarized in Table 1. In the overall cohort, 55.2% (629/1139) were male, and the median age at diagnosis was 59.00 years. The median follow‐up duration was 26.6 months. ISS Stages II and III were observed in 35.3% (364/1030) and 40.2% (414/1030) of patients, respectively. IgG was the most common M protein subtype (47.4%, 535/1128), followed by IgA (24.6%, 277/1128) and light chain (21.3%, 240/1128).

FIGURE 1.

Patients disposition. MM, multiple myeloma; SMM, smoldering multiple myeloma; NDMM, newly diagnosed multiple myeloma; HIS, hospital information system.

TABLE 1.

Patient characteristics at baseline in overall and high‐risk population.

Characteristic	All n = 1139	Age			Renal impairment			1q abnormalities#
		≥ 65 years n = 364	< 65 years n = 775	p	Renal impairment n = 148	Without renal impairment n = 418	p	With n = 334	Without n = 339	p
Sex, male, n (%)	629 (55.2)	211 (58.0)	418 (53.9)	0.225	80 (54.1)	216 (51.7)	0.687	174 (52.1)	192 (56.6)	0.269
Age, years, median (IQR)	59.0 (53.0, 66.0)	69.0 (67.0, 72.0)	55.0 (50.0, 60.0)	< 0.001	61.0 (53.0, 67.5)	59.0 (52.0, 65.0)	0.013	60.0 (54.0, 66.0)	59.0 (52.0, 66.0)	0.125
Age by category, years, n (%)							0.034			0.585
< 65	775 (68.0%)	NA	775 (100.0%)		94 (63.5%)	306 (73.2%)		225 (67.4%)	236 (69.6%)
≥ 65	364 (32.0%)	364 (100.0%)	NA		54 (36.5%)	112 (26.8%)		109 (32.6%)	103 (30.4%)
Length of follow up, months, median (IQR)	26.6 (18.2, 39.4)	24.7 (17.6, 36.1)	28.0 (18.8, 40.7)	0.009	24.8 (17.2, 35.8)	28.4 (18.7, 40.7)	0.054	27.6 (19.2, 40.7)	29.1 (21.3, 42.0)	0.162
ISS stage, n (%)*				< 0.001			< 0.001			0.006
I	252 (24.5%)	65 (19.5%)	187 (26.9%)		3 (2.2%)	111 (29.3%)		53 (17.8%)	84 (26.5%)
II	364 (35.3%)	105 (31.4%)	259 (37.2%)		9 (6.7%)	162 (42.7%)		106 (35.6%)	121 (38.2%)
III	414 (40.2%)	164 (49.1%)	250 (35.9%)		123 (91.1%)	106 (28.0%)		139 (46.6%)	112 (35.3%)
Missing	109	30	79		13	39		36	22
M‐protein isotype, n (%)*				0.081			0.004			< 0.001
IgA	277 (24.6%)	92 (25.6%)	185 (24.1%)		33 (22.4%)	100 (24.2%)		102 (30.6%)	70 (20.8%)
IgD	48 (4.3%)	9 (2.5%)	39 (5.1%)		8 (5.4%)	18 (4.3%)		15 (4.5%)	7 (2.1%)
IgG	535 (47.4%)	186 (51.7%)	349 (45.4%)		51 (34.7%)	206 (49.8%)		163 (48.9%)	166 (49.3%)
IgG + IgA	4 (0.4%)	0 (0.0%)	4 (0.5%)		0 (0.0%)	0 (0.0%)		1 (0.3%)	2 (0.6%)
IgM	3 (0.3%)	2 (0.6%)	1 (0.1%)		0 (0.0%)	1 (0.2%)		2 (0.6%)	1 (0.3%)
Light chain	240 (21.3%)	65 (18.1%)	175 (22.8%)		52 (35.4%)	82 (19.8%)		49 (14.7%)	82 (24.3%)
Non‐secreting	20 (1.8%)	6 (1.7%)	14 (1.8%)		3 (2.0%)	7 (1.7%)		1 (0.3%)	9 (2.7%)
Oligosecretory	1 (0.1%)	0 (0.0%)	1 (0.1%)		0 (0.0%)	0 (0.0%)		0 (0.0%)	0 (0.0%)
Missing	11	4	7		1	4		1	2

Note: Bold values indicate statistical significance.

Abbreviations: IQR, interquartile; ISS, International Staging System.

^# 1q abnormality analysis based on available FISH data.

^*Analysis was conducted based on non‐missing data.

3.2. Frontline Treatment Patterns

In the frontline setting, the most commonly used regimens in the overall population were PI+IMiD‐based combinations (50.0%), followed by PI‐based regimens (38.6%). Treatment selection varied significantly by age (p = 0.001); elderly patients received PI+IMiD‐based regimens less frequently than younger patients (45.3% vs. 52.3%), whereas PI‐based regimens (40.1% vs. 37.9%) and IMiD‐based regimens (6.0% vs. 1.9%) were more frequently used in elderly patients. Treatment patterns differed significantly by renal function status (p < 0.001), with patients with baseline RI more often receiving PI‐based regimens (46.6% vs. 28.2%), while those without RI more commonly received PI+IMiD‐based regimens (60.3% vs. 36.5%). Notably, treatment patterns were generally similar regardless of 1q abnormality status. A total of 92 (8.1%) patients in the overall population received a CD38‐based regimen as part of their initial induction (Table 2).

TABLE 2.

Frontline treatment patterns in overall and subgroups.

Characteristic	All n = 1139	Age < 65 years n = 775	Age ≥ 65 years n = 364	With renal impairment n = 148	Without renal impairment n = 418	With 1q abnormalities n = 334	Without 1q abnormalities n = 339
Treatment patterns, n (%)
CD38‐based	92 (8.1%)	61 (7.9%)	31 (8.5%)	19 (12.8%)	41 (9.8%)	22 (6.6%)	18 (5.3%)
PI+IMiD‐based	570 (50.0%)	405 (52.3%)	165 (45.3%)	54 (36.5%)	252 (60.3%)	175 (52.4%)	169 (49.9%)
PI‐based	440 (38.6%)	294 (37.9%)	146 (40.1%)	69 (46.6%)	118 (28.2%)	130 (38.9%)	144 (42.5%)
IMiD‐based	37 (3.3%)	15 (1.9%)	22 (6.0%)	6 (4.1%)	7 (1.7%)	7 (2.1%)	8 (2.4%)
p		0.001		< 0.001		0.741
Transplantation, n (%)
ASCT	286 (25.1%)	261 (33.7%)	25 (6.9%)	25 (16.9%)	128 (30.6%)	84 (25.2%)	87 (25.7%)
p		< 0.001		0.002		0.948

Note: Bold values indicate statistical significance.

Abbreviations: ASCT, autologous stem cell transplantation; CD38, CD38 monoclonal antibodies; IMiD, immunomodulatory drug; PI, proteasome inhibitor.

Overall, 25.1% of patients underwent ASCT as part of their frontline treatment. The proportion of patients receiving ASCT was markedly lower among elderly individuals (6.9% vs. 33.7%) and those with RI (16.9% vs. 30.6%). No significant sex‐related differences were observed in frontline regimen selection (Table S1).

3.3. Frontline Treatment Patterns Among Patients Received ASCT/Non‐ASCT

The PI+IMiD‐based regimen was the most commonly used treatment in the overall population and in all subgroups except those with renal impairment. Among patients who underwent transplantation, elderly patients were more likely than younger patients to receive a CD38‐based regimen (16.0% vs. 8.4%). Patients with RI were more likely to receive a PI‐based regimen (40.0% vs. 19.5%) compared to those without renal impairment. This observation likely reflects treatment selection among transplant‐eligible patients, in whom clinicians may preferentially use PI‐based regimens in the setting of RI to achieve rapid disease control before transplantation. Patients with 1q abnormalities were more likely to receive a PI+IMiD‐based regimen than those with normal 1q status (69.1% vs. 59.8%).

Among patients who did not undergo transplantation, elderly patients were more likely than younger patients to receive an IMiD‐based regimen (6.2% vs. 2.9%). Patients with RI were more likely to receive a CD38‐based regimen than those without RI (13.0% vs. 8.6%). Treatment patterns were generally similar between patients with 1q abnormalities and those with normal 1q status (Figure 2).

FIGURE 2.

Frontline treatment patterns among patients with different transplant status. (A) patients received ASCT; (B) patients did not undergo ASCT. CD38, CD38 monoclonal antibodies; PI, proteasome inhibitor; IMiD, immunomodulatory drug; ASCT, autologous stem cell transplantation.

3.4. Subsequent Treatment Pattern After First Relapse

Among 304 patients experiencing documented disease progression, treatment patterns after first relapse differed from those used in the frontline setting (Figure 3). In the overall population, the use of CD38‐based regimens increased from 5.3% during frontline treatment to 30.9% after first relapse. PI+IMiD‐based regimens remained commonly used (43.8% vs. 32.6%), whereas the use of PI‐based regimens decreased from 45.7% to 16.1%. In contrast, IMiD‐based regimens and conventional regimens became more frequently used after relapse (5.3% vs. 16.5% and 0% vs. 4.0%, respectively). Similar treatment shifts were observed across elderly patients, patients with renal impairment, and patients with 1q abnormalities. Among patients who experienced disease relapse during the observation period, 8.9% (27/304) received ASCT as part of second‐line treatment.

FIGURE 3.

Frontline and subsequent treatment after first relapse. CD38, CD38 monoclonal antibodies; PI, proteasome inhibitor; IMiD, immunomodulatory drug.

3.5. Clinical Outcomes (PFS and ORR)

In the frontline setting, PFS was estimated using the Kaplan–Meier method. In the overall NDMM population, the median PFS was 46.8 months. The estimated 12‐month and 24‐month PFS rates were 93.2% (95% CI: 91.4%–95.0%) and 78.2% (95% CI: 75.0%–81.7%), respectively (Figure 4A). When stratified by age, patients aged ≥ 65 years demonstrated numerically lower 12‐ and 24‐month PFS rates compared with those aged < 65 years (Figure 4B). Similarly, although early PFS estimates were comparable between patients with and without renal impairment, a gradual separation of the Kaplan–Meier curves was observed over time (Figure 4C). Patients with 1q abnormalities also showed shorter PFS compared with those without 1q abnormalities (Figure 4D).

FIGURE 4.

The estimated progression‐free survival (PFS) probability in frontline treatment. (A) overall population; (B) Age < 65 years vs. Age ≥ 65 years; (C) With renal impairment vs. Without renal impairment; (D) With 1q abnormalities vs. Without 1q abnormalities.

Although the ORR during frontline treatment remained around 80% across all groups, the proportion of patients achieving a VGPR or better ranged only between 60.9% and 75.1% (Figure 5). In the overall NDMM population, the ORR during frontline treatment was 81.00%, but only 67.37% of patients achieved a VGPR or better response. Although the ORR was similar between the ≥ 65 years group and the < 65 years group, the ≥ 65 years group had a notably lower proportion of patients achieving VGPR or better (60.9% vs. 70.5%). A similar pattern was observed for deeper responses, with a lower proportion of patients achieving CR or better in elderly patients (34.9% vs. 51.1%). Patients without RI had significantly higher ORR (92.0% vs. 83.7%) and VGPR or better response rates (75.1% vs. 68.3%) compared to those with renal impairment. Although the ORR was comparable between patients with and without 1q abnormalities (84.2% vs. 83.4%), the proportion achieving VGPR or better was lower in the 1q abnormalities group (65.0% vs. 71.8%).

FIGURE 5.

Treatment responses in overall and high‐risk subgroups. SCR, stringent complete response; CR, complete response; VGPR, very good partial response; PR, partial response; MR, minor response; SD, stable disease; PD, progressive disease; NE, no evidence.

Exploratory Cox regression analyses were performed to evaluate baseline factors associated with PFS. In the overall cohort, 514 patients with complete data for the included covariates were included in the multivariable analysis, among whom 94 PFS events were observed. ISS Stage II and ISS Stage III were independently associated with shorter PFS. In the corresponding analysis restricted to patients with available baseline FISH data, 340 patients with 70 PFS events were included in the multivariable model. ISS Stage II and ISS Stage III remained independently associated with shorter PFS, whereas baseline RI (HR [95% CI]: 1.71 [1.02–2.85]; p = 0.040) and 1q abnormalities (HR [95% CI]: 1.51 [0.94–2.42]; p = 0.089) showed associations with inferior PFS in univariable analyses (Table S2).

In the separate exploratory cytogenetic subgroup analysis, 594 patients with 137 PFS events were included in the multivariable model. Patients were categorized as without 1q abnormalities (n = 339), isolated 1q abnormalities (n = 245), or 1q abnormalities co‐occurring with additional HRCAs (n = 89). The median PFS was 58.6 months in patients without 1q abnormalities, 40.3 months in patients with isolated 1q abnormalities, and 42.0 months in patients with 1q abnormalities plus additional HRCAs (Figure S1). In the multivariable analysis, the HRs were 1.37 (95% CI, 0.95–1.98; p = 0.095) for isolated 1q abnormalities and 1.73 (95% CI, 1.00–3.00; p = 0.049) for 1q abnormalities plus additional HRCAs, compared with patients without 1q abnormalities (Table S3).

In the sensitivity analysis using the same covariate framework and complete‐case population as the FISH‐restricted baseline‐factor model, ISS Stage II (HR [95% CI]: 2.37 [1.06–5.32]; p = 0.036) and ISS Stage III (HR [95% CI]: 2.97 [1.28–6.91]; p = 0.012) remained independently associated with shorter PFS. The HRs were 1.17 (95% CI, 0.70–1.95; p = 0.559) for isolated 1q abnormalities and 1.68 (95% CI, 0.76–3.72; p = 0.200) for 1q abnormalities plus additional HRCAs (Table S4).

4. Discussion

4.1. Treatment Patterns and Novel Agents

With the integration of novel agents and ASCT in the clinical management of MM, treatment patterns for NDMM patients have evolved, resulting in marked improvements in patient outcomes. This multicenter, retrospective, observational study (CHAMP‐MM) investigated the real‐world treatment patterns and outcomes among Chinese patients with NDMM. Our findings suggest that PI‐ and IMiD‐based regimens are now widely used in China, providing substantial therapeutic benefits. Since the inclusion of agents like bortezomib, lenalidomide, and daratumumab in the national medical insurance system in 2016, PI+IMiD‐based combinations have emerged as the preferred frontline option. In our cohort, all patients received either a PI or an IMiD, with nearly 10% receiving a CD38‐based regimen following the regulatory approval of daratumumab for frontline use in 2021.

Upon first relapse, there was a notable increase in the utilization of CD38‐based regimens compared to the frontline setting. Concurrently, the use of IMiD‐based and conventional regimens increased, particularly among elderly patients—a trend likely reflecting the prior use of PI‐containing options. Notably, the adoption of PI+IMiD‐based regimens has risen significantly in recent years compared with historical data, highlighting a clear evolution toward more intensive frontline standards in Chinese real‐world practice [12, 15, 16].

4.2. ASCT

Only 25.11% of patients underwent ASCT, and this rate was significantly lower among elderly patients. Although ASCT utilization has gradually increased in China [17, 18, 19, 20], it remains lower than in developed countries [21]. Financial constraints may be a contributing factor, as out‐of‐pocket costs still pose a barrier despite ASCT being covered by national insurance. In addition, the low penetration rate of ASCT technology may also limit its widespread use. The implementation of ASCT promotion and education programs to improve awareness and acceptance of ASCT among transplant‐eligible patients remains important.

Elderly patients were less likely to undergo ASCT, likely due to poorer health status and ineligibility. Concerns regarding the safety and tolerability of ASCT in older populations also influence clinical decision‐making [7]. Similarly, patients with RI showed a lower rate of ASCT and PI+IMiD use. Although lenalidomide dosing can be adjusted for renal function [22], clinicians often hesitate to use it before renal recovery. Patients with RI may benefit from ASCT in terms of efficacy but are at higher risk for transplant‐related mortality [23].

Compared to patients who did not undergo ASCT, those who did were more likely to receive triplet regimens or higher‐intensity combinations (PI+IMiD or CD38‐based) and were typically younger (< 65 years) with fewer comorbidities [24].

4.3. High‐Risk Subgroups and Specific Observations

This study included a high proportion of elderly patients (≥ 65 years), those with renal impairment, and those with 1q abnormalities—subgroups often underrepresented in clinical trials. These patients differed from the general cohort in treatment selection and ASCT eligibility, which subsequently impacted their clinical outcomes.

Consistent with previous study in Chinese NDMM [19], our study showed a long mPFS during frontline treatment, likely attributable to the widespread adoption of novel agent‐based therapies [25]. Although the ORR (PR or better: 81%) to frontline therapy was favorable, the proportion of patients achieving deeper responses (VGPR or better: 67.37%) was modest, possibly due to the low ASCT rate. Meanwhile, patients with RI demonstrated inferior PFS over time compared to those without renal impairment, which may be partly attributed to the lower proportion of these patients receiving PI+IMiD‐based or more intensive regimens during frontline therapy. Patients with 1q abnormalities had treatment patterns similar to the patients without 1q abnormalities but showed worse PFS. This findings may indicate that currently used treatment strategies do not fully overcome the adverse prognostic impact associated with high‐risk cytogenetic features [26, 27]. In exploratory Cox regression analyses, ISS Stage II and ISS Stage III were consistently associated with shorter PFS in both the overall cohort and the FISH‐available subgroup. In the separate exploratory cytogenetic analysis, patients with 1q abnormalities co‐occurring with additional HRCAs showed a higher estimated risk of disease progression than patients without 1q abnormalities, whereas isolated 1q abnormalities showed a weaker association. However, in the sensitivity analysis using the same complete‐case population and covariate framework as the FISH‐restricted baseline‐factor model, the effect estimates for ISS stage remained consistent, while the association between 1q abnormalities plus additional HRCAs and PFS was no longer statistically significant. Although the point estimate for this cytogenetic subgroup remained similar in direction and magnitude, these findings indicate that the statistical significance of the 1q‐related association was sensitive to analytic sample size and model specification. Therefore, the prognostic implications of the exploratory cytogenetic subgroup analysis should be interpreted cautiously.

4.4. Limitations and Future Directions

This study has several limitations. As a retrospective study, data collection was limited for certain clinical variables such as copy number in FISH testing, and minimal residual disease (MRD) status. As a result, this study was unable to explore whether deeper remission (e.g., MRD negative) is associated with longer disease control, or whether different types of 1q abnormalities (e.g., gain (1q) vs. amp (1q)) have distinct impacts on clinical outcomes. Additionally, the follow‐up duration was relatively short; only 26.7% of all patients had experienced relapse during the study period, likely due to the prolonged treatment course of MM and the widespread use of novel agents, which has transformed MM into a more chronic, long‐term managed disease. However, although progression events were sufficient for PFS analyses, the number of observed death events during follow‐up was very limited. As a result, a meaningful evaluation of overall survival was not feasible at this stage despite the availability of mortality information. In addition, as follow‐up was based on routine clinical visits in real‐world practice, the frequency and timing of assessments may have varied according to patients' clinical condition. Therefore, informative censoring cannot be entirely excluded, which may have influenced time‐to‐event estimates. Future prospective studies with longer follow‐up and more granular clinical data are needed to validate these findings and further explore optimized strategies for high‐risk subgroups.

5. Conclusion

In conclusion, this study provides a comprehensive overview of real‐world treatment patterns and clinical outcomes of NDMM patients in China over the past 5 years. Our findings demonstrate that PI+IMiD‐based regimens have been established as the backbone of frontline therapy, and the utilization of CD38‐based regimens has increased among patients who relapsed after frontline therapy. Across the overall clinical outcome analyses, including treatment response and progression‐free survival, elderly patients, those with renal impairment, and those with 1q abnormalities generally experienced less favorable clinical outcomes. These data underscore the need for tailored therapeutic strategies and intensified management protocols to address these persisting clinical gaps.

Author Contributions

Jian Cui: conceptualization, data curation, visualization, writing – original draft, writing – review and editing. Xiaoyan Qu: data curation, resources. Chunrui Li: data curation, resources. Fei Li: data curation, resources. Liye Zhong: data curation, resources. Xiaohong Fu: conceptualization, formal analysis, supervision, writing – original draft, writing – review and editing. Gang Liu: conceptualization, formal analysis, supervision, writing – original draft, writing – review and editing. Lugui Qiu: conceptualization, data curation, project administration, funding acquisition, supervision, resources, writing – original draft, writing – review and editing. Gang An: conceptualization, data curation, project administration, funding acquisition, supervision, resources, writing – original draft, writing – review and editing.

Funding

This study was funded by Sanofi. J.C. was supported by the National Natural Science Foundation of China (824B2004). G.A. was supported by the National Natural Science Foundation of China (82670295, U22A20291, and 82270218), the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (2025‐I2M‐C&T‐A‐013), the Tianjin Natural Science Foundation Project (25JCLMJC01120), and the Noncommunicable Chronic Diseases–National Science and Technology Major Project (2023ZD0501300).

Ethics Statement

The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Science and Peking Union Medical College (Approval No. QT2024001‐EC‐1). All other participating centers also obtained approval from their respective institutional ethics committees prior to data collection. The ethical approval numbers for the participating centers were as follows: Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University (2024‐MD‐102), Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology ([2024]S065), The First Affiliated Hospital, Jiangxi Medical College, Nanchang University (IIT[2024]166), and The First Affiliated Hospital of Jinan University (KY‐2024‐174). The requirement for individual patient consent was waived due to the retrospective nature of the study, as approved by the ethics committees.

Consent

Informed consent was waived due to the retrospective nature of the study.

Conflicts of Interest

Xiaohong Fu, Gang Liu are Sanofi employees and may hold stocks of Sanofi. No potential conflicts of interest to disclose from other authors.

Supporting information

Figure S1: The estimated progression‐free survival probability in frontline treatment according to 1q abnormality status among patients with available FISH data. FISH, fluorescence in situ hybridization; HRCA, high‐risk cytogenetic abnormalities.

Table S1: Frontline treatment patterns stratified by sex.

Table S2: Univariable and multivariable Cox regression analyses of factors associated with PFS in the overall cohort and in patients with available baseline FISH data.

Table S3: Multivariable Cox regression analysis of progression‐free survival according to cytogenetic subgroup among patients with available FISH data.

Table S4: Sensitivity multivariable Cox regression analysis of progression‐free survival according to cytogenetic subgroup among patients with available baseline FISH data.

Data Availability Statement

The dataset generated and analyzed during the current study are not publicly available due to data privacy regulations and ethical restrictions. However, de‐identified data may be accessed through a formal request to the corresponding author. Qualified researchers may obtain access to the data for non‐commercial, academic purposes, subject to approval by the study team and in compliance with relevant data protection and sharing regulations.