Sage Journals: Discover world-class research

Abstract

Introduction

The introduction of proteasome inhibitors and immunomodulatory agents has significantly improved the prognosis of multiple myeloma (MM). However, the occurrence of second primary malignancies (SPMs) in MM survivors has raised widespread concern.

Methods

This population-based retrospective study using the SEER database analyzed data from 26,869 MM patients (1990–2021) to evaluate changes in SPMs across two therapeutic eras. Patients were stratified into two therapeutic eras based on the year of MM diagnosis: Era-1 (1990–2005; n=12,858) and Era-2 (2006–2021; n=14,011).

Results

Among 1,346 MM patients who developed SPMs, 670 were in Era-1 and 676 in Era-2. The 15-year cumulative incidence of SPMs was significantly higher in Era-2 (7.7% vs. 4.8%, P < 0.001), an increase driven mainly by solid tumors (6.90% vs. 4.10%, P < 0.001) with no significant change in second hematological malignancies (0.84% vs. 0.67%, P = 0.13). Standardized incidence ratio (SIR) analysis revealed elevated hematological malignancy risk in Era-2 (SIR = 1.71, 95% CI: 1.49–1.96). Median time to SPM was shorter in Era-2 (43.5 vs. 59 months, P < 0.001). Notably, 80% of SPMs in Era-2 occurred within 90 months of MM diagnosis. Overall survival (OS) of SPM patients showed no significant improvement in Era-2 compared to Era-1. Within each diagnostic era, patients who developed SPMs exhibited longer overall survival than those with MM alone; however, this reflects survivor bias, as SPMs can only develop in patients who survive long enough after the initial MM diagnosis. No significant OS differences were observed among SPM patients by race or gender.

Conclusions

The risk of SPMs in MM survivors has significantly increased, and the latency between MM diagnosis and the onset of SPMs was shorter in the Era-2, highlighting the need for enhanced cancer surveillance in MM survivors.

Keywords

multiple myeloma second primary malignancies therapeutic era cumulative incidence prognosis

Introduction

Multiple myeloma (MM) is a common plasma cell malignancy, accounting for approximately 10% of all hematological malignancies. MM exhibits a slightly higher incidence in men than in women, with a median age at diagnosis of approximately 65 years.^1,2 Over the past two decades, the introduction of proteasome inhibitors such as bortezomib and immunomodulatory agents (IMiDs) including thalidomide and lenalidomide has significantly improved survival outcomes in MM patients, extending the median survival from 3–5 to 7–10 years.³ However, with prolonged overall survival (OS), the development of second primary malignancies (SPMs) following MM treatment has emerged as a critical clinical challenge.

IMiDs, including lenalidomide and thalidomide, exert their anti-myeloma activity through a variety of mechanisms. Specifically, these agents induce cell cycle arrest and apoptosis in MM cells⁴; downregulate adhesion molecule expression, thereby reducing MM cell binding to both cellular components and extracellular matrix proteins within the bone marrow microenvironment⁵; inhibit angiogenesis⁶; modulate cytokine production and secretion⁷; enhance the activity of T cells and natural killer cells, while modulating the function of regulatory T cells.^8-10 However, the immunomodulatory properties of IMiDs, while critical for anti-myeloma efficacy, may also contribute to an increased risk of SPMs in patients with MM. Studies have found that immunosuppressive therapy can reactivate the EBV lytic cycle of quiescent B lymphocytes and lead to immunosuppression.¹¹ Another potential effect of lenalidomide on B cells is the alteration of the CRL4 E3 ubiquitin ligase complex through the cereblon protein, i.e., the selective ubiquitination and degradation of the IKAROS protein (encoded by IKZF1) and AIOLOS protein (encoded by IKZF3).^12,13 IKAROS and AIOLOS transcription factors are critical for the regulation of B-cell function. Lenalidomide can also amplify regulatory T cells (Tregs) to induce immune tolerance.¹⁴ Multiple studies suggested a potential rise in the incidence of SPMs after lenalidomide treatment. In the large phase III CALGB 100104 trial, lenalidomide maintenance after autologous stem-cell transplantation in newly diagnosed multiple myeloma patients was linked to a higher SPM risk compared with placebo. Invasive SPMs occurred in 32 patients (13.9%) in the lenalidomide group (18 hematological [7.8%] and 14 solid tumors [6.1%]) versus 12 patients (5.2%) in the placebo group (3 hematological [1.3%] and 9 solid tumors [3.9%]).¹⁵ Similarly, the multicenter IFM 2005-02 trial reported a higher SPM incidence in patients receiving lenalidomide maintenance post-transplantation (3.1 per 100 patient-years vs. 1.2 in the placebo group; P=0.002).¹⁶ In the Myeloma XI trial, a randomized study in newly diagnosed patients, the 3-year cumulative incidence of SPMs was 5.3% (95% CI 3.6–7.1) in the lenalidomide group versus 3.1% (95% CI 1.8–4.5) in the observation group (HR 1.85, 95% CI 1.18–2.90).¹⁷ However, studies assessing the change in SPM risk before and after the introduction of IMiDs and proteasome inhibitors in 2006 are limited, and no population-based study has yet comprehensively evaluated the survival outcomes of SPM patients across both treatment eras.

This study focused on comparing the incidence, tumor type, prognostic outcomes, and demographic characteristics of SPMs in MM survivors before and after the introduction of proteasome inhibitors and IMiDs. Importantly, we also explored the critical window for monitoring SPMs in MM survivors during the new treatment era. Using the Surveillance, Epidemiology, and End Results (SEER) database, we identified 26,869 patients diagnosed with primary MM between 1990 and 2021 across eight U.S. regions, among whom 1,346 subsequently developed SPMs. We reported the cumulative incidence, SIRs, latency periods, cancer subtypes, demographic and clinical characteristics, and survival outcomes associated with SPMs. To our knowledge, this represents the largest cohort study to date assessing SPMs among MM patients, spanning nearly three decades and encompassing two therapeutic eras: Era-1 (1990–2005), prior to the introduction of IMiDs and proteasome inhibitors, and Era-2 (2006–2021), following their adoption into clinical practice. The findings of this study provide novel insights into the risk of SPMs in MM patients during the immunochemotherapy era, contributing to enhanced cancer surveillance in MM survivors, and potentially guiding the development of future treatment strategies.

Materials and Methods

Database and Participants

This population-based retrospective observational study utilized routinely collected data from the 8 original SEER registries (SEER-8), covering the period from January 1, 1975, to December 31, 2021. Additionally, a subset of patients diagnosed between January 1, 1990, and December 31, 2021, was further included for analysis. Primary tumor sites were classified using the International Classification of Diseases for Oncology, Third Edition (ICD-O-3), with inclusion criteria restricted to patients pathologically confirmed as “9732/3: Plasma cell myeloma”. Patients were selected consecutively based on all histologically confirmed cases meeting the above criteria during the study period. Exclusion criteria included individuals with MM as a second malignancy and patients lacking complete data on radiotherapy, chemotherapy, age, sex, race, survival status, or follow-up information. This study utilized publicly available, de-identified data from the SEER program, which does not require ethics approval. As this research involved the analysis of existing public data, individual consent to participate was not required. This study reported the cohort in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. The reporting of this study conforms to the RECORD statement.¹⁸

Definition and Follow-up of SPMs

SPMs were defined as any subsequent malignant neoplasm (ICD-O-3 behavior code/3) recorded in the SEER database with a “Multiple Primary Fields – Sequence Number” value indicating “2nd of 2 or more primaries” (i.e., the second or later primary malignancy in the patient’s lifetime) and for which multiple myeloma was the first recorded primary malignancy. This sequence number and the corresponding multiple-primary designation are assigned automatically by the SEER program only after full application of the official SEER Multiple Primary and Histology Coding Rules (including site-specific, histology-specific, and timing rules). Consequently, all SPMs included in this analysis strictly adhere to the internationally recognized IARC/IACR/SEER criteria for multiple primary cancers, thereby excluding recurrences, metastases, or disease transformations that do not qualify as new primary malignancies.^19-22

Hematological SPMs were defined using the following ICD-O-3 morphology codes. Leukemia included codes 9823, 9827, 9831–9837, 9840, 9860–9861, 9863, 9865-9867, 9869–9876, 9891, 9895–9898, 9910–9911, 9920, 9930–9931, 9940, 9945-9946, 9948. Non-Hodgkin lymphoma (both nodal and extranodal) included codes 9591, 9670–9671, 9673, 9675, 9678–9680, 9684, 9687–9691, 9695, 9698–9719, 9726-9729, 9735, 9737-9738, 9761-9762.

The latency to SPM was defined as the interval from the date of MM diagnosis to the date of diagnosis of the subsequent primary malignancy. To minimize ascertainment bias due to intensified diagnostic workup at the time of MM diagnosis and to exclude true synchronous malignancies, only SPMs diagnosed ≥6 months after the MM diagnosis were included in the analysis. This minimum 6-month latency requirement is consistent with the approach used in numerous large-scale population-based cancer registry studies.²³

To comprehensively assess SPM risk, overall risks for all SPMs, solid tumors, and hematological malignancies were first calculated. Subsequent risk analyses were performed separately for predefined organ sites, including prostate, colon, breast, lung and bronchus, leukemia, and non-Hodgkin lymphoma (NHL). Additionally, subgroup analyses were conducted stratified by radiotherapy exposure to evaluate its potential influence on SPM cumulative incidence across therapeutic eras. Follow-up for SPMs commenced 6 months after MM diagnosis and ceased at the earlier of SPM diagnosis or all-cause death. The final follow-up date was designated as December 31, 2021.

Statistical Analysis

To evaluate the impact of modern therapies on survival, patients were classified into two diagnostic eras based on their MM diagnosis year: 1990–2005 (Era-1) and 2006–2021 (Era-2). The year 2006 was selected as the dividing line between the two therapeutic eras because it corresponds to the U.S. FDA approval of lenalidomide (Revlimid®) in combination with dexamethasone for patients with multiple MM who had received at least one prior therapy. This regulatory milestone represented a turning point in MM management. In the pivotal trials supporting approval, lenalidomide plus dexamethasone significantly improved overall response rate (60.6% vs. 21.9%, P<0.001), complete response rate (15.0% vs. 2.0%, P<0.001), median time to progression (13.4 months vs. 4.6 months, P<0.001), and duration of response (15.8 months vs. 7 months, P<0.001) compared with placebo plus dexamethasone,²⁴ thereby establishing a new standard of care in the relapsed/refractory setting and rapidly facilitating its incorporation into frontline and maintenance strategies.^25,26

Cumulative incidence of SPMs was assessed via Fine-Gray competing risk regression analysis. Competing risks for SPM development comprised all-cause mortality and occurrence of non-SPM events. Categorical variables were compared using chi-squared tests, with Fisher’s exact test applied when expected cell frequencies were fewer than five. The Mann-Whitney U test evaluated continuous variables regardless of normality. Statistical analyses were performed using R software (version 4.4.1; R Project for Statistical Computing). The SIR was defined as the ratio of observed SPM incidence among MM survivors to that in the general US population. SIRs were calculated using SEER*Stat software (version 8.4.5). In the analysis, these ratios were adjusted for age at MM diagnosis and year of diagnosis. To assess the prognosis of patients with SPMs, the Kaplan-Meier approach was used to estimate OS, with P values derived from log-rank testing. OS was defined as the time from MM diagnosis to death from any cause. Stratified analyses by age, sex, and race were conducted to assess prognostic characteristics across SPM patient subgroups. Patients with documented race/ethnicity were classified into non-Hispanic Whites (White), non-Hispanic Blacks (Black), Hispanic Whites (Hispanic), and non-Hispanic Asians or Pacific Islanders (Asian/PI). Patients with Native American or Alaska Native race were excluded owing to insufficient numbers for meaningful statistical assessment (n=12).

Results

Patient Characteristics

From 1990 to 2021, a total of 32,768 MM cases were identified in the SEER database (Supplementary Figure 1). Following data validation, the final analytic cohort comprised 26,869 patients with primary multiple myeloma, of whom 12,858 (47.9%) were diagnosed in Era-1 (1990–2005) and 14,011 (52.1%) in Era-2 (2006–2021). In the cohort, 18,531 patients (69%) were White, with a median age of 68 years at MM diagnosis (interquartile range, 59–76 years). The male-to-female ratio was approximately 1.2:1, and 15,650 patients (58%) were married at the time of diagnosis. In terms of treatment, 17,731 patients (66%) received chemotherapy, and 5,730 patients (21%) had radiation therapy. Across both therapeutic eras, a total of 1,346 patients (5.0% of the entire cohort) developed at least one SPM during follow-up, comprising 670 patients (5.2%) in Era-1 and 676 patients (4.8%) in Era-2. The baseline characteristics of patients who developed SPMs are detailed in Supplementary Table 1.

Cumulative Incidences and Latency of SPMs

In Era-2, the 15-year cumulative incidence of combined SPMs after MM diagnosis was 7.7%, compared with 4.8% in Era-1 (P < 0.001) (Figure 1A). This overall increase was driven predominantly by second solid tumors (6.90% vs. 4.10%, P < 0.001; Figure 1B). Notably, the cumulative incidence of second hematological malignancies remained essentially unchanged between the two eras (0.84% vs. 0.67%, P = 0.13; Figure 1C), indicating that the era-related rise in SPMs was confined mainly to solid tumors.

Figure 1.

The comparisons of cumulative incidence of the top 10 second primary malignancies (SPMs). (A) All SPMs combined (B) solid cancer (C) hematologic malignancies (D) prostate cancer (E) large intestine cancer (F) breast cancer (G) lung and bronchus cancer (H) miscellaneous (I) leukemia (J) melanoma of the skin (K) urinary bladder cancer (L) NHL and (M) kidney and renal pelvis cancer. P values were calculated with the Fine-Gray test. Detailed ICD-O-3 morphology codes for leukemia and NHL are provided in the Materials and methods section

We further examined the ten most frequent SPM sites across both eras combined (ranked regardless of solid or hematological origin). Among these, the five sites that demonstrated significantly higher 15-year cumulative incidence in Era-2 were all solid tumors: prostate cancer (1.50% vs. 0.77%, P < 0.001), lung and bronchus cancer (0.70% vs. 0.38%, P = 0.002), urinary bladder cancer (0.49% vs. 0.23%, P = 0.006), melanoma of the skin (0.46% vs. 0.23%, P = 0.006), and kidney and renal pelvis cancer (0.44% vs. 0.09%, P = 0.007). No significant era-related differences were observed for the remaining top sites, including large intestine cancer, breast cancer, leukemia, non-Hodgkin lymphoma, and miscellaneous cancer (Figure 1D–M).

Radiotherapy was recorded in 21% of the overall cohort (5,730/26,869 patients), with significantly higher utilization in Era-1 (25%, 3,188/12,858) than in Era-2 (18%, 2,542/14,011; P < 0.001) (Table 1). Subgroup analysis stratified by radiotherapy exposure revealed that the increase in SPM cumulative incidence in Era-2 persisted regardless of radiotherapy status. Among patients who received radiotherapy, the 15-year cumulative incidence of SPMs was 7.2% in Era-2 compared with 3.7% in Era-1 (P < 0.001) (Supplementary Figure 2A). Similarly, among patients who did not receive radiotherapy, the 15-year cumulative incidence was 7.8% in Era-2 versus 5.1% in Era-1 (P < 0.001) (Supplementary Figure 2B). Within Era-1, patients who did not receive radiotherapy had a higher 15-year cumulative incidence than those who did (5.1% vs. 3.7%, P = 0.006) (Supplementary Figure 2C). In Era-2, no significant difference was observed between irradiated and non-irradiated patients (7.2% vs. 7.8%, P = 0.71) (Supplementary Figure 2D). Across the overall cohort, non-irradiated patients exhibited a slightly higher 15-year cumulative incidence (7.8% vs. 5.1%, P = 0.022) (Supplementary Figure 2E).

Table 1.

Comparisons of Baseline Characteristics of Patients With MM by Treatment Era

Characteristic	Overall	Era-1	Era-2	p-value
	Overall	Before 2006	After 2006
	N = 26,869	N = 12,858	N = 14,011
Age at MM diagnosis, median	68 (59-76)	69 (59-77)	68 (59-76)	<0.001
Age at MM Diagnosis				<0.001
<50	2,243 (8.3%)	1,093 (8.5%)	1,150 (8.2%)
50-65	8,225 (31%)	3,720 (29%)	4,505 (32%)
65-80	11,847 (44%)	5,811 (45%)	6,036 (43%)
80+	4,554 (17%)	2,234 (17%)	2,320 (17%)
Sex				0.051
Female	12,209 (45%)	5,922 (46%)	6,287 (45%)
Male	14,660 (55%)	6,936 (54%)	7,724 (55%)
Race/ethnicity				<0.001
Hispanic	2,033 (7.6%)	725 (5.6%)	1,308 (9.3%)
Indian/Alaska Native	248 (0.9%)	111 (0.9%)	137 (1.0%)
Asian/PI	2,081 (7.7%)	858 (6.7%)	1,223 (8.7%)
Black	3,976 (15%)	1,612 (13%)	2,364 (17%)
White	18,531 (69%)	9,552 (74%)	8,979 (64%)
Marital				<0.001
Married	15,650 (58%)	7,647 (59%)	8,003 (57%)
Single/unmarried	9,415 (35%)	4,512 (35%)	4,903 (35%)
Unknown	1,804 (6.7%)	699 (5.4%)	1,105 (7.9%)
Chemotherapy for MM				<0.001
No/Unknown	9,138 (34%)	4,154 (32%)	4,984 (36%)
Yes	17,731 (66%)	8,704 (68%)	9,027 (64%)
Radiation for MM				<0.001
Beam radiation	5,730 (21%)	3,188 (25%)	2,542 (18%)
None/Unknown	21,139 (79%)	9,670 (75%)	11,469 (82%)
Patients who developed SPMs	1,346 (5.0%)	670 (5.2%)	676 (4.8%)	0.15

For continuous variables, P - values were calculated using the Mann - Whitney test. For categorical variables, P - values were calculated using the χ2 test.

The median latency from multiple myeloma diagnosis to SPM was significantly shorter in Era-2 than in Era-1 (43.5 vs. 59 months; Figure 2A). Strikingly, 80% of all SPMs in Era-2 were diagnosed within 90 months of the initial myeloma diagnosis, compared with 145 months in Era-1 (Figure 2B–C). This marked left-shift in the latency distribution indicates that the vast majority of SPMs now occur relatively early in the disease course of contemporary patients.

Figure 2.

The distribution of the latency time for SPMs. (A) The latency time for SPM in MM patients between Era-1 and Era-2 era. (B) 80% area under the curve of the latency time for patients developing SPM in the Era-1 (Before 2006). (C) 80% area under the curve of the latency time for patients developing SPM in the Era-2 (After 2006)

Subgroup analysis stratified by SPM type revealed a consistent pattern across both therapeutic eras: median latency to solid-tumor SPMs was shorter than to hematological SPMs, whereas the 80th percentile latency was longer for solid tumors in both eras, reflecting differences in distribution shapes (Supplementary Figure 3). In the overall cohort, median latency was 49 months for solid tumors versus 58 months for hematological SPMs; 80% of solid-tumor SPMs occurred within 112 months, compared with 114 months for hematological SPMs (Supplementary Figure 3A–C). In Era-1 (1990–2005), median latency was 57 months for solid tumors versus 68.5 months for hematological SPMs; 80% of solid-tumor SPMs occurred within 149 months, compared with 142 months for hematological SPMs (Supplementary Figure 3D–F). In Era-2 (2006–2021), median latency was 43 months for solid tumors versus 46 months for hematological SPMs; 80% of solid-tumor SPMs occurred within 92 months, compared with 88 months for hematological SPMs (Supplementary Figure 3G–I).

Standardized Incidence Ratios (SIRS) of SPMs

SIRs for all solid tumors, hematolymphoid malignancies, and the top 10 organ systems identified in prior analyses are reported in Table 2. Overall, the SIR of all hematologic malignancies among MM survivors was significantly increased in Era-2 (SIR = 1.71, 95% CI: 1.49–1.96), while no significant difference was observed for all solid tumors. At the organ-system level, elevated SIRs were also observed for malignancies of the colon/rectum/anus, miscellaneous, leukemia, and kidney/renal pelvis. By contrast, no significant changes in SIRs between the two eras were detected for prostate, breast, lung/bronchus, melanoma, bladder cancer, or non-Hodgkin lymphoma.

Table 2.

Standardized Incidence Ratios (SIR) With 95% Confidence Intervals (CIs) of SPMs After an Initial Diagnosis of MM by Treatment Era, 1990–2021

Site of the SPMs	Era-1 (1990-2005)	Era-2 (2006-2021)
Site of the SPMs	SIR (95% CIs)	SIR (95% CIs)
All Solid Tumors	0.89(0.82-0.96)	0.97(0.92-1.03)
All Lymphatic and Hematopoietic Diseases	1.37(1.09-1.70)	1.71(1.49-1.96)
Prostate	0.68(0.56-0.82)	0.87(0.76-1.00)
Colon, Rectum and Anus	1.03(0.84-1.26)	1.26(1.06-1.48)
Breast	0.83(0.64-1.06)	0.75(0.62-0.91)
Lung and Bronchus	0.76(0.61-0.94)	0.87(0.74-1.02)
Miscellaneous	1.36(0.93-1.92)	1.78(1.44-2.18)
Leukemia	1.95(1.38-2.67)	2.71(2.23-3.27)
Melanoma of the Skin	1.32(0.87-1.92)	1.14(0.90-1.43)
Urinary Bladder	1.08(0.78-1.45)	0.95(0.75-1.20)
NHL	1.37(0.97-1.88)	1.06(0.81-1.36)
Kidney and Renal Pelvis	1.17(0.72-1.79)	1.77(1.40-2.20)

SIRs were calculated using SEER*Stat software (version 8.4.5).

SPMs Profile in Two Eras

The bar plot illustrates the 20 most common types of SPMs (Figure 3). Among the 1,346 SPM cases identified in our study cohort, the most frequently observed malignancy was prostate cancer (n=234), accounting for 17.8% of all SPMs and 28.9% of male SPM cases. This was followed by large intestine cancer (n=174, 13.2%), breast cancer (n=120, all: 9.1%, female: 23.2%), lung and bronchus cancer (n=114, 8.6%), and miscellaneous (n=80, 6%). The types of SPMs in MM patients remained largely unchanged between the two eras; however, the proportions of specific SPM types differed. Incidences of Kidney and Renal Pelvis cancer (n=12 vs n=36, P < 0.001) and thyroid cancer (n=7 vs n=19, P = 0.03) were higher in Era-2 compared to Era-1.

Figure 3.

SPM subtypes in all patients 1990–2021. * indicates P values <0.05, P values were calculated with the χ2 tests

Survival Outcome of SPMs

We compared OS among MM patients across different treatment eras. No survival difference was observed between MM patients diagnosed with SPMs in Era-2 and Era-1 (P = 0.74) (Figure 4A). By contrast, the first primary MM patients in Era-2 exhibited a significant OS advantage compared with Era-1 patients (P < 0.001) (Figure 4B). Within the same treatment era, SPM patients exhibited improved OS compared to those with MM alone (all P < 0.001) (Figure 4C–D).

Figure 4.

Overall survival (OS) Kaplan–Meier curves of patients with SPMs. (A) Kaplan-Meier survival curve comparing OS between SPM patients in two eras. (B) Kaplan-Meier survival curve comparing OS between MM patients with and without SPMs in two eras. (C) Kaplan-Meier survival curve comparing OS between SPM patients and MM patients without SPMs during the Era-1. (D) Kaplan-Meier survival curve comparing OS between SPM patients and MM patients without SPMs during the Era-2. P-values were calculated using the log-rank test

In the age-stratified analysis, SPM patients in the 65–80 years subgroup demonstrated superior OS in Era-2 compared with Era-1. And no significant OS differences were observed between eras in patients aged <50 years, 50–65 years, or >80 years (Supplementary Figure 4A–D). Analysis across four major racial groups (White, Hispanic, Black, and Asian/PI) revealed no significant differences in OS among SPM patients by race (Supplementary Figure 4E–F). When evaluating OS by race and age group, a survival advantage for Era-2 SPM patients was observed exclusively in White patients aged 65–80 years (P = 0.001) (Supplementary Table 2). SPM patients in Era-2 with Hispanic ethnicity diagnosed at 65–80 years exhibited a survival disadvantage (P = 0.029). Black and Asian/PI patients showed no significant OS differences across all age groups. Among male and female patients, no significant survival differences were observed between eras for those who developed SPMs (Supplementary Figure 5A–B). Within all SPM patients, OS did not differ significantly between males and females (Supplementary Figure 5C–D). In Era-2, male SPM patients of all racial groups showed no significant OS differences compared with Era1 (Supplementary Table 3). Among females, a survival disadvantage in Era-2 was observed only in Hispanic patients.

Discussion

To our knowledge, this is the largest population-based study to date evaluating the evolving patterns of SPMs in MM patients before and after the introduction of proteasome inhibitors and IMiDs. The key findings of this study are as follows: First, patients with MM in Era-2 exhibited a significantly higher cumulative incidence and SIR of SPMs compared with those in Era-1. The latency to SPM development was shorter in Era-2, with 80% of SPMs diagnosed within 7.5 years following the initial MM diagnosis. Second, the increase in SPMs incidence was predominantly driven by solid tumors. Notably, in Era-2, the proportions of kidney/renal pelvis and thyroid cancers increased significantly, while the distribution of other SPM subtypes remained largely stable. Third, patients with MM in Era-2 demonstrated significantly longer OS compared to those in Era-1; however, among patients who developed SPMs, no significant difference in OS was observed between the two eras.

In our analysis, MM survivors in Era-2 exhibited a significantly elevated risk of developing SPMs compared with those in Era-1, accompanied by a shortened latency from MM diagnosis to SPM onset. In the modern treatment era, the incidence of SPMs among cancer survivors has shown a notable increase compared to earlier periods. A previous population-based study also reported an increased risk of second hematologic malignancies among childhood cancer survivors, particularly second acute myeloid leukemia (AML), with a rising trend observed from 1986 to 2005.²⁷ Another study compared the risk of SPMs among chronic lymphocytic leukemia survivors identified from the SEER database (1973–2015) with the expected incidence of various malignancies in the general population. Over approximately 270,000 person-years of follow-up, 6,487 new SPMs were identified, yielding a SIR of 1.2 (95% CI: 1.17–1.23).²⁸ A case-control study from Korea further demonstrated that patients with MM had a significantly increased risk of developing hematologic malignancies compared to the general population.²⁹ However, limited studies have assessed the changes in SPM risk across treatment eras defined by the introduction of proteasome inhibitors and IMiDs. In the present study, multiple myeloma survivors diagnosed in Era-2 (2006–2021) exhibited a higher cumulative incidence of SPMs compared with those diagnosed in Era-1 (1990–2005). This observation coincides temporally with the widespread adoption of novel agents (proteasome inhibitors and immunomodulatory drugs) and the substantial improvement in overall survival achieved during the same period. Accumulating evidence from malignancies has demonstrated that the use of specific anti-cancer agents, including radiotherapy, alkylating agents, and topoisomerase II inhibitors, is associated with an increased risk of SPMs.³⁰

We found that patients with MM in Era-2 exhibited a significantly higher cumulative incidence and SIR of SPMs compared with those in Era-1. To further explore the role of traditional treatment factors, we conducted subgroup analyses stratified by radiotherapy status. The results showed that, despite a significant decline in radiotherapy utilization in Era-2 (Era-1 vs Era-2:25% vs 18%,P < 0.001), the Era-2 increase in SPM cumulative incidence persisted in both irradiated and non-irradiated patients. The observed elevation in SPM cumulative incidence during Era-2 was evident across both radiated and non-radiated subgroups, suggesting that radiotherapy is not the primary driver of the increased SPM rates in the modern therapeutic era. This finding is consistent with the declining utilization of radiotherapy in MM management, as systemic therapies such as proteasome inhibitors and IMiDs have become predominant, potentially contributing to SPM risks through mechanisms like genomic instability or immunosuppression.

Similarly, the relationship between novel agents and SPM risk remains controversial. Early studies reported no clear association between the use of novel agents (including thalidomide and lenalidomide) and SPM risk in patients with multiple myeloma.^31-33 However, these findings were limited by relatively short follow-up durations. Lenalidomide was approved for the treatment of MM in 2006. In a randomized trial involving 460 patients with a median follow-up of 91 months, lenalidomide maintenance therapy following autologous stem cell transplantation (ASCT) was associated with a significantly increased risk of SPMs compared to placebo.¹⁵ A meta-analysis further demonstrated that lenalidomide treatment for MM was associated with increased risks of both solid and hematologic SPMs, in both transplant and non-transplant settings.³⁴ A retrospective study of Chinese patients with MM reported a 2.99% incidence of SPMs (2/67) following bortezomib monotherapy.³⁵ Overall, although novel agents—including thalidomide, bortezomib, and lenalidomide—have dramatically improved survival outcomes in multiple myeloma, their potential relationship with SPM development remains controversial and inconclusive. The same era has also witnessed a paradigm shift from conventional cytotoxic chemotherapy toward prolonged induction, consolidation, and maintenance regimens that routinely incorporate these novel agents. However, the SEER database does not collect individual-level data on specific drug classes, doses, treatment duration, combination regimens, autologous stem-cell transplantation, or maintenance therapy. Consequently, we are unable to establish direct causality between any particular agent or regimen and the observed increase in SPM incidence. The era-related associations observed in this study may primarily reflect the combined influence of factors such as prolonged patient survival and intensified cancer surveillance, rather than a specific carcinogenic effect attributable to novel agents. Complementary studies with detailed treatment records are required to disentangle these competing explanations.

This study demonstrated a significant reduction in SPM latency in Era-2. The median latency was reduced from 59 to 43.5 months, and approximately 80% of SPMs, mainly solid tumors, developed within 90 months following myeloma diagnosis, versus 145 months in Era-1. The observed shortening of median latency to SPMs in Era-2, despite prolonged survival, may reflect era-specific shifts in treatment paradigms and surveillance practices. The introduction of novel agents predominant in Era-2, such as immunomodulatory drugs and proteasome inhibitors, has been hypothesized in prior studies to potentially accelerate SPM onset through mechanisms including genomic instability, impaired DNA repair, and promotion of clonal hematopoiesis, though direct causal links remain unestablished in population-based data like SEER due to the absence of treatment details. Additionally, advancements in MM management during Era-2, including enhanced cancer surveillance with more frequent imaging and laboratory monitoring amid improved overall health outcomes, could facilitate earlier detection of SPMs. Our findings highlight the need for a targeted surveillance window encompassing approximately the first 7.5 years (90 months) following MM diagnosis in the contemporary era. Intensified monitoring, particularly for solid tumors—which primarily drove the increased SPM incidence—during this period may facilitate early detection and timely intervention of SPMs.

The increase in overall SPM cumulative incidence observed in the novel-agent era was driven mainly by solid tumors. The five solid tumor sites with statistically significantly higher 15-year cumulative incidence in Era-2 compared with Era-1 were: prostate, lung and bronchus, urinary bladder, melanoma of the skin, and kidney and renal pelvis cancers. Prostate cancer remained the most common SPM in both eras, consistent with prior population-based studies.³⁶ No significant era-related increases were observed for other major solid tumors (including colorectal, breast, thyroid, and pancreatic cancers).

In marked contrast, the cumulative incidence of second hematological malignancies remained stable between the two eras, despite earlier reports from clinical trials suggesting an increased risk associated with certain novel agents.³⁴ This discrepancy may be explained by the low absolute event numbers (limiting statistical power), reduced alkylating-agent exposure in modern risk-adapted strategies, and the absence of detailed cytogenetic or treatment data in SEER. Nonetheless, the unchanged incidence of hematological SPMs amid substantially improved myeloma survival provides population-level reassurance, indicating that any leukemogenic effect of novel agents, if present, is modest compared with the predominant contribution of solid tumors to the overall SPM burden.

We used cumulative incidence and SIR as complementary metrics for SPM risk. Cumulative incidence reflects the absolute probability of SPM in the presence of competing risks and is sensitive to prolonged survival, whereas SIR measures relative excess risk versus an age-, sex-, and calendar-year-matched general population. The increased cumulative incidence but unchanged SIR for several solid tumors (e.g., prostate and lung cancer) in Era-2 suggests that the higher absolute burden is driven predominantly by extended survival rather than therapy-specific oncogenicity, with these malignancies influenced mainly by shared factors such as aging, smoking, and chronic immunosuppression. In contrast, persistently elevated SIR for hematological SPMs (leukemia and non-Hodgkin lymphoma; SIR 1.37–1.71) despite stable event numbers reflects their low background incidence in the general population, indicating a modest but lingering leukemogenic potential associated with multiple myeloma and its historical treatments. Clinically, cumulative incidence is the more actionable metric for guiding surveillance timing and intensity in contemporary myeloma survivors, while SIR is primarily useful for etiological inference.

A key focus of this study was the impact of SPMs on long-term survival in multiple myeloma. While overall survival in the general myeloma population improved substantially in the novel-agent era (Era-2) due to proteasome inhibitors and immunomodulatory drugs, patients who developed SPMs in Era-2 showed no corresponding prolongation in survival after SPM diagnosis compared with Era-1 counterparts. This lack of benefit is attributable to two main factors: first, the predominant SPMs in Era-2 were aggressive solid tumors with poor prognosis (particularly lung cancer),³⁷ whose outcome is driven primarily by the second malignancy itself, negating benefits from improved myeloma therapy; second, the longer initial survival in Era-2 meant SPMs typically occurred after multiple lines of therapy, often in the context of relapsed/refractory disease and cumulative organ impairment, limiting tolerance to intensive second-cancer treatment. Consequently, advances in myeloma management have not translated into improved post-SPM survival. These findings highlight the critical need for early detection and aggressive management of solid tumors in long-term myeloma survivors.

Interestingly, within each diagnostic era, patients who developed SPMs exhibited longer overall survival than those with myeloma alone. This apparent advantage, however, is largely attributable to survivor bias: SPMs can only be diagnosed in patients who have survived sufficiently long after initial myeloma diagnosis (median latency 43.5 months in Era-2 and 59 months in Era-1) for a second cancer to manifest. Consequently, the SPM cohort is enriched for long-term survivors with inherently better prognosis from the outset. This selection artefact prolongs observed survival in the SPM group without implying any true benefit from the second malignancy or a favorable prognostic impact of SPMs.

Although we primarily used 2006—the year of lenalidomide approval—as the therapeutic era breakpoint to capture widespread adoption of IMiDs, alternative cutoffs merit brief consideration. Bortezomib was initially approved in 2003 with an indication restricted to patients with relapsed or refractory multiple myeloma who had received at least two prior therapies³⁸; however, substantial survival benefits at the population level were primarily achieved following its combination with IMiDs after 2006. Similarly, definitive evidence for lenalidomide maintenance post-autologous transplantation emerged in 2010–2012 trials,^16,39 leading to gradual adoption thereafter. Thus, 2003 predates broad novel-agent synergy, while 2010 represents refinements within an established paradigm. The 2006 breakpoint therefore best delineates the foundational shift to highly effective novel-agent regimens and provides the optimal framework for assessing associations between contemporary myeloma therapies and SPM risk.

The strengths of this study include its large, population-based cohort and long follow-up, providing adequate power to detect era-related changes in SPM patterns among multiple myeloma survivors. We reported both cumulative incidence (accounting for competing risks) and standardized incidence ratios to comprehensively assess SPM risk across eras. However, like all SEER-based analyses, a major limitation is the lack of individual-level treatment data, preventing drug- or regimen-specific conclusions on oncogenicity. These findings therefore require validation in cohorts with detailed therapeutic records. Additionally, as the study is based on U.S. SEER registry data, its generalizability to non-U.S. populations with differing healthcare systems, treatment practices, or genetic backgrounds may be limited. Furthermore, survival comparisons between patients with and without SPMs are susceptible to survivor bias, as the SPM cohort is inherently selected for longer initial survival sufficient for a second malignancy to develop. Finally, subgroup analyses for specific SPM types or sites may be underpowered due to relatively low event numbers, despite the large overall cohort size.

Conclusions

In conclusion, our study results indicate that after the introduction of proteasome inhibitors and IMiDs, the incidence of SPMs in MM survivors may have significantly increased, accompanied by changes in SPM types. Notably, we observed that the latency period for SPMs in MM survivors has shortened in the new treatment era. These observations may be related to broader treatment approaches in the current era. In terms of survival prognosis, the OS of the SPM patient group showed no significant improvement compared to Era-1.

Supplemental Material

Supplemental Material - The Changing Landscape of Second Primary Malignancies in Multiple Myeloma: A SEER Population-Based Study Between Two Therapeutic Eras

Supplemental Material for The Changing Landscape of Second Primary Malignancies in Multiple Myeloma: A SEER Population-Based Study Between Two Therapeutic Eras by Weixiang Lu, Xiaoshan Huang, Shengyu Tian, Xiaojie Liang, Guanzhou Ke, Jia Guo, Qingqing Li, Yuquan Huang, Yang Li, Baiwei Luo, Bingyu Lin, Dan Xu, Liang Wang in Cancer Control.

Footnotes

Acknowledgements

The authors are grateful to the SEER database for providing the data that facilitated this research.

ORCID iDs

Weixiang Lu

Xiaoshan Huang

Liang Wang

Ethics Considerations

Not applicable. This study utilized publicly available, de-identified data from the Surveillance, Epidemiology, and End Results (SEER) program, which does not require ethics approval.

Consent to Participate

Not applicable. As this research involved the analysis of existing public data, individual consent to participate was not required.

Consent for Publication

Not applicable. This study does not contain any individual person’s data in any identifiable form. The data analyzed in this research are aggregated and publicly available, ensuring privacy and confidentiality are maintained.

Author Contributions

Conceptualization: Dan Xu, Liang Wang; Methodology: Dan Xu, Liang Wang; Formal Analysis: Weixiang Lu, Xiaoshan Huang, Shengyu Tian; Investigation: Guanzhou Ke; Data Curation: Guanzhou Ke; Visualization: Weixiang Lu, Xiaoshan Huang, Shengyu Tian; Writing—Original Draft Preparation: Weixiang Lu, Xiaoshan Huang; Writing—Review & Editing: Shengyu Tian, Xiaojie Liang, Jia Guo, Qingqing Li, Yuquan Huang, Yang Li, Baiwei Luo, Binyu Lin, Dan Xu, Liang Wang; Funding Acquisition: Dan Xu, Liang Wang. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Beijing High-Level Innovation and Entrepreneurship Talent Support Program leading talent projects (G202512029) to Liang Wang.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data supporting the findings of this study are available from the Surveillance, Epidemiology, and End Results (SEER) program database, which is publicly accessible for research purposes. Due to the public nature of the data, all analyses are based on previously published data; this study does not contain any new data collected by the authors.*

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Rajkumar

. Multiple myeloma: 2022 update on diagnosis, risk stratification, and management. Am J Hematol. 2022;97(8):1086-1107. doi:10.1002/ajh.26590.

Cowan

Green

Kwok

, et al. Diagnosis and Management of Multiple Myeloma: A Review. Jama. 2022;327(5):464-477. doi:10.1001/jama.2022.0003.

Bobin

Liuu

Moya

, et al. Multiple Myeloma: An Overview of the Current and Novel Therapeutic Approaches in 2020. Cancers. 2020;12(10):2885. doi:10.3390/cancers12102885.

Hideshima

Chauhan

Shima

, et al. Thalidomide and its analogs overcome drug resistance of human multiple myeloma cells to conventional therapy. Blood. 2000;96(9):2943-2950.

Gupta

Treon

Shima

, et al. Adherence of multiple myeloma cells to bone marrow stromal cells upregulates vascular endothelial growth factor secretion: therapeutic applications. Leukemia. 2001;15(12):1950-1961. doi:10.1038/sj.leu.2402295.

D'Amato

Loughnan

Flynn

Folkman

. Thalidomide is an inhibitor of angiogenesis. Proceedings of the National Academy of Sciences of the United States of America. 1994;91(9):4082-4085. doi:10.1073/pnas.91.9.4082.

Corral

Haslett

Muller

, et al. Differential cytokine modulation and T cell activation by two distinct classes of thalidomide analogues that are potent inhibitors of TNF-alpha. J Immunol. 1999;163(1):380-386.

Chang

Liu

Klimek

, et al. Enhancement of ligand-dependent activation of human natural killer T cells by lenalidomide: therapeutic implications. Blood. 2006;108(2):618-621. doi:10.1182/blood-2005-10-4184.

Zhu

Corral

Fleming

Stein

. Immunomodulatory drugs Revlimid (lenalidomide) and CC-4047 induce apoptosis of both hematological and solid tumor cells through NK cell activation. Cancer Immunol Immunother. 2008;57(12):1849-1859. doi:10.1007/s00262-008-0512-7.

10.

Galustian

Meyer

Labarthe

, et al. The anti-cancer agents lenalidomide and pomalidomide inhibit the proliferation and function of T regulatory cells. Cancer Immunol Immunother. 2009;58(7):1033-1045. doi:10.1007/s00262-008-0620-4.

11.

Jones

Iempridee

Wang

, et al. Lenalidomide, Thalidomide, and Pomalidomide Reactivate the Epstein-Barr Virus Lytic Cycle through Phosphoinositide 3-Kinase Signaling and Ikaros Expression. Clin Cancer Res. 2016;22(19):4901-4912. doi:10.1158/1078-0432.Ccr-15-2242.

12.

Krönke

Udeshi

Narla

, et al. Lenalidomide causes selective degradation of IKZF1 and IKZF3 in multiple myeloma cells. Science (New York, NY). 2014;343(6168):301-305. doi:10.1126/science.1244851.

13.

Middleton

Sun

, et al. The myeloma drug lenalidomide promotes the cereblon-dependent destruction of Ikaros proteins. Science (New York, NY). 2014;343(6168):305-309. doi:10.1126/science.1244917.

14.

Balaian

Schuster

Schönefeldt

, et al. Selective expansion of regulatory T cells during lenalidomide treatment of myelodysplastic syndrome with isolated deletion 5q. Ann Hematol. 2016;95(11):1805-1810. doi:10.1007/s00277-016-2775-y.

15.

Holstein

Jung

Richardson

, et al. Updated analysis of CALGB (Alliance) 100104 assessing lenalidomide versus placebo maintenance after single autologous stem-cell transplantation for multiple myeloma: a randomised, double-blind, phase 3 trial. Lancet Haematol. 2017;4(9):e431-e442. doi:10.1016/s2352-3026(17)30140-0.

16.

Attal

Lauwers-Cances

Marit

, et al. Lenalidomide maintenance after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366(19):1782-1791. doi:10.1056/NEJMoa1114138.

17.

Jackson

Davies

Pawlyn

, et al. Lenalidomide maintenance versus observation for patients with newly diagnosed multiple myeloma (Myeloma XI): a multicentre, open-label, randomised, phase 3 trial. Lancet Oncol. 2019;20(1):57-73. doi:10.1016/s1470-2045(18)30687-9.

18.

Benchimol

Smeeth

Guttmann

, et al. The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement. PLoS Med. 2015;12(10):e1001885. doi:10.1371/journal.pmed.1001885.

19.

Anderson

Mayer

Nichols

. Trends in the proportion of second or later primaries among all newly diagnosed malignant cancers. Cancer. 2021;127(15):2736-2742. doi:10.1002/cncr.33558.

20.

Murphy

Gerber

Pruitt

. Prevalence of Prior Cancer Among Persons Newly Diagnosed With Cancer: An Initial Report From the Surveillance, Epidemiology, and End Results Program. JAMA Oncol. 2018;4(6):832-836. doi:10.1001/jamaoncol.2017.3605.

21.

Johnson

CPS

Adamo

Fritz

Percy-Laurry

Edwards

. The 2007 multiple primary and histology coding rules. National Cancer Institute, Surveillance, Epidemiology and End Results Program; 2007.

22.

International rules for multiple primary cancers (ICD-0 third edition). Eur J Cancer Prev. 2005;14(4):307-308. doi:10.1097/00008469-200508000-00002.

23.

Okines

Thomson

Radstone

Horsman

Hancock

. Second primary malignancies after treatment for malignant lymphoma. Br J Cancer. 2005;93(4):418-424. doi:10.1038/sj.bjc.6602731.

24.

Dimopoulos

Chen

Spencer

, et al. Long-term follow-up on overall survival from the MM-009 and MM-010 phase III trials of lenalidomide plus dexamethasone in patients with relapsed or refractory multiple myeloma. Leukemia. 2009;23(11):2147-2152. doi:10.1038/leu.2009.147.

25.

McCaughan

Gandolfi

Moore

Richardson

. Lenalidomide, bortezomib and dexamethasone induction therapy for the treatment of newly diagnosed multiple myeloma: a practical review. Br J Haematol. 2022;199(2):190-204. doi:10.1111/bjh.18295.

26.

Falco

Cavallo

Larocca

, et al. Lenalidomide and its role in the management of multiple myeloma. Expert Rev Anticancer Ther. 2008;8(6):865-874. doi:10.1586/14737140.8.6.865.

27.

Rihani

Bazzeh

Faqih

Sultan

. Secondary hematopoietic malignancies in survivors of childhood cancer: an analysis of 111 cases from the Surveillance, Epidemiology, and End Result-9 registry. Cancer. 2010;116(18):4385-4394. doi:10.1002/cncr.25313.

28.

Kumar

Ailawadhi

Bojanini

, et al. Trends in the risk of second primary malignancies among survivors of chronic lymphocytic leukemia. Blood Cancer J. 2019;9(10):75. doi:10.1038/s41408-019-0237-1.

29.

Han

Park

Choi

Byun

Min

. Risk of Secondary Malignancies After Multiple Myeloma: A Nationwide Case-Control Cohort Study. Clin Lymphoma Myeloma Leuk. 2024;24(10):e366-e375. doi:10.1016/j.clml.2024.05.015.

30.

Morton

Swerdlow

Schaapveld

, et al. Current knowledge and future research directions in treatment-related second primary malignancies. EJC Suppl. 2014;12(1):5-17. doi:10.1016/j.ejcsup.2014.05.001.

31.

Mailankody

Pfeiffer

Kristinsson

, et al. Risk of acute myeloid leukemia and myelodysplastic syndromes after multiple myeloma and its precursor disease (MGUS). Blood. 2011;118(15):4086-4092. doi:10.1182/blood-2011-05-355743.

32.

Razavi

Rand

Cozen

Chanan-Khan

Usmani

Ailawadhi

. Patterns of second primary malignancy risk in multiple myeloma patients before and after the introduction of novel therapeutics. Blood Cancer J. 2013;3(6):e121. doi:10.1038/bcj.2013.19.

33.

Rifkin

Abonour

Shah

, et al. Connect MM® - the Multiple Myeloma Disease Registry: incidence of second primary malignancies in patients treated with lenalidomide. Leuk Lymphoma. 2016;57(9):2228-2231. doi:10.3109/10428194.2015.1132419.

34.

Saleem

Franz

Klem

, et al. Second primary malignancies in patients with haematological cancers treated with lenalidomide: a systematic review and meta-analysis. Lancet Haematol. 2022;9(12):e906-e918. doi:10.1016/s2352-3026(22)00289-7.

35.

Guo

Huang

Wen

. Two Secondary Primary Malignancies after Bortezomib Therapy for Multiple Myeloma: A Single-center Experience. Chin Med J (Engl). 2017;130(2):239-241. doi:10.4103/0366-6999.198017.

36.

Avivi

Vesole

Davila-Valls

, et al. Outcome of Second Primary Malignancies Developing in Multiple Myeloma Patients. Cancers. 2023;15(17):15. doi:10.3390/cancers15174359.

37.

Sung

Ferlay

Siegel

, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71(3):209-249. doi:10.3322/caac.21660.

38.

Rajkumar

Richardson

Hideshima

Anderson

. Proteasome inhibition as a novel therapeutic target in human cancer. J Clin Oncol. 2005;23(3):630-639. doi:10.1200/jco.2005.11.030.

39.

McCarthy

Owzar

Hofmeister

, et al. Lenalidomide after stem-cell transplantation for multiple myeloma. N Engl J Med. 2012;366(19):1770-1781. doi:10.1056/NEJMoa1114083.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.02 MB