Abstract
The 2/20/20 International Myeloma Working Group (IMWG) score is the most employed risk score in clinical practice to evaluate the risk of progression from smoldering multiple myeloma (SMM) to symptomatic multiple myeloma. However, it faces a serious limitation: The risk score is applied at diagnosis and cannot be reapplied. Since a dynamic accurate patient risk assessment for progression is necessary, we aimed to investigate whether the detection of an evolving pattern in serum M‐protein (SMP) improves the identification of high‐risk patients. Eighty‐three patients diagnosed with SMM between 2011 and 2020 were included. Patients were initially classified applying the 2/20/20 IMWG score at baseline and later reclassified depending on the presence of an SMP evolving pattern into six groups. We regrouped the patients into three final risk groups: low‐risk, intermediate‐risk, and high‐risk. The risk of progression at two years for the high‐risk group was 88% and all patients had progressed at 4 years. The performance measurements were superior for the new 2/20/20‐Evolving score independently for the detection of high‐risk patients. We show that the sequential measurement of the SMP is a noninvasive and widely available test that improves the 2/20/20 IMWG risk score.
INTRODUCTION
Smoldering multiple myeloma (SMM) is an asymptomatic monoclonal gammopathy with a higher tumor burden than the monoclonal gammopathy of unknown significance (MGUS) but without the presence of multiple myeloma (MM) defining events. It is characterized by the presence of a serum monoclonal protein (SMP) ≥ 3 g/dL in serum and/or ≥500 mg/24 h in urine and/or ≥10% bone marrow plasma cells (BMPC), in the absence of end‐organ damage (CRAB features: hypercalcemia, renal insufficiency, anemia, or bone lesions). 1
SMM constitutes a heterogeneous clinicopathological entity with some patients showing an indolent course (“MGUS‐like”) while others have a more aggressive behavior with a rapid progression to MM (“early MM”). 2 , 3 Retrospective studies have shown that the progression rate (PR) to MM is variable over time: 10% per year during the first 5 years, 3% per year for the next 5 years, and 1% thereafter. 4
The International Myeloma Working Group (IMWG) updated in 2014 the diagnostic criteria for SMM to exclude those patients with ultra‐high‐risk features, due to studies that showed an 80% risk of progression at 2 years. 5 These ultra‐high‐risk features include clonal BMPC > 60%, a serum free‐light chain ratio (FLCr) > 100, and >1 focal lesion on whole‐body magnetic resonance imaging (MRI). These patients are now considered to have active MM and treatment should be offered.
Due to the heterogeneity of the course of the disease, several prognostic scales have been developed to define the risk of progression to MM. The Mayo Clinic published a risk stratification model in 2018 using the IMWG 2014 diagnostic criteria, which was validated by the IMWG in 2020, called the 2/20/20 model. It includes the SMP, BMPC, and FLCr as risk factors and allows to optionally include cytogenetic data if available. 6
Another widely used score was developed by the PETHEMA group. This score considers the presence of two adverse factors: ≥95% of BMPC with abnormal immunophenotype and immunoparesis. 7
All these scores are applied at diagnosis and do not take into account that some of the variables described above follow a dynamic pattern. In this sense, Rosiñol's group identified the increase in SMP over time as a new risk factor for progression and described the evolving type of SMM, characterized by a progressive increase in the SMP size and a shorter time to progression (TTP) to symptomatic MM, probably reflecting different pathogenetic mechanisms. 3 Moreover, it was suggested that an evolving SMP (defined as a ≥10% increase in SMP within 12 months from diagnosis if baseline SMP ≥3 g/dL or within 3 years if baseline SMP <3 g/dL) was an independent risk factor of progression to MM, with a significantly shorter TTP to symptomatic disease. 8
The objective of this study is to explore whether the incorporation of the SMP evolving pattern to the IMWG (2/20/20) score allows for a more accurate identification of patients at high risk of progressing to symptomatic MM.
MATERIALS AND METHODS
Patients
Clinical records of patients diagnosed with SMM between 2011 and 2020 in a single hospital center were reviewed. Patients who met the SLiM criteria were excluded. 5 The Ethics Committee provided institutional review board approval for this study.
Diagnosis of SMM was established according to the IMWG criteria 5 : ≥10% BMPCs and/or SMP ≥3 g/dL plus absence of hypercalcemia (calcium >11.5 mg/dL), renal failure (creatinine >2 mg/dL), anemia (hemoglobin <10 g/dL), or bone lesions attributable to the gammopathy, start of therapy for MM in the absence of CRAB features, or development of systemic light‐chain amyloidosis.
Outcome measures
TTP was calculated as the length of time from the diagnosis of SMM to the start of antimyeloma therapy due to the development of symptomatic disease.
Prognostic risk scores
The risk of progression to symptomatic MM was assessed using the IMWG 2/20/20 risk score (excluding the cytogenetic analysis), the PETHEMA risk model, and the presence of an evolving pattern.
The 2/20/20 IMWG score assesses three risk factors for progression: SMP >2 g/dL, BMPC >20%, and FLCr >20 as cut‐offs, 6 leading to three risk categories with increasing 2‐year progression risk from 6% for low risk (no risk factors above the cut‐off) to 18% for intermediate risk (1 risk factor) and 44% for high risk (2 or more risk factors). 6
The PETHEMA risk model 7 confers one point to the presence of ≥95% aberrant plasma cells in bone marrow using multiparameter flow cytometry and one point to the presence of immunoparesis, defined as uninvolved immunoglobulin serum level below the normal range. This risk model also identifies three risk groups with progression‐free survival (PFS) at 5 years: 4% for low‐risk (0 points), 46% for intermediate risk (1 point), and 72% for high‐risk group (2 points).
The evolving versus nonevolving pattern of SMP was evaluated by serum electrophoresis at least every 3 months for 1 year. The evolving pattern was considered when a progressive increase of SMP size by at least 10% was observed in those with a baseline SMP of 3 g/dL or higher while a 25% increase was required for those with a baseline SMP below 3 g/dL, always confirmed in a second measurement. 8 , 9 The cut‐off point was determined by two methods: based on previous published studies and statistical methodology. First, the used threshold was initially described by Rosiñol et al. (British Journal of Haematology, 2003) when defining the “evolving” pattern for the first time and was later replicated by Fernandez de Larrea et al. (Leukemia, 2018) and Ravi et al. (Blood Cancer Journal, 2016). Second, the maxstat and Cutoff‐Gray codes in R were used to confirm this series, and a similar result was obtained (Supporting Information S1: Figure 1).
Statistical analysis
Continuous data were summarized as median and interquartile range (IQR). Categorical data were represented as proportions. Statistical comparisons of categorical variables were performed by using the Chi‐square test and Fisher's test, meanwhile, for continuous variables, the t‐test or Mann–Whitney U test was used. The evolving pattern was analyzed as a time‐dependent covariate. 10
Time‐to‐event analyses were done with Gray's test for competing risk. The primary outcome was progression to symptomatic MM. Patients were censored if they did not progress at the date of the last follow‐up or started therapy within a clinical trial for SMM. The competing event was death without progression. The Fine‐Gray proportional hazard regression with a time‐dependent covariate model was used to elucidate the impact of different risk scores on TTP. The evolving pattern was added to the 2/20/20 IMWG risk model, resulting in a total of six new risk categories. The categories were then regrouped according to the risk of progression using a statistical approach, thus creating the new prognostic score. The median and the 2‐ and 4‐year TTP were calculated for all risk models. The median follow‐up of the population was calculated using the reverse Kaplan–Meier method. Graphical representation of time‐dependent covariates with two levels was performed by the Simon–Makuch method. 11 Time‐dependent receiver operating characteristics (ROC) curves, area under the curve (AUC), and concordance index (C‐index) analyses were used to determine the performance characteristics of the risk scores. The C‐index confirms the predictive power of the assessed score for values greater than 0.5 with perfect discrimination of the value of 1. 12 Diagnostic accuracy measures (sensitivity, specificity, and predictive values) were calculated at 2 years. To make clinical decisions that may provide benefits over alternative decision criteria, a decision curve analysis was performed. This method is used to determine whether the use of a prediction model in the clinic to inform decision‐making would be better than detrimental. 13 The Sankey plot was used to depict the dynamic change in patient risk from the diagnosis of SMM to the detection of the evolving pattern. All analyses were evaluated using a two‐tailed P value with an alpha level of 0.05. Analyses were performed using R 4.3.0 (R Foundation for Statistical Computing, http://www.R-project.org) and EZR 1.61.
RESULTS
Patient characteristics
Ninety‐three patients were diagnosed with SMM in our institution between 2011 and 2020. Ten of them were not included in the study: nine due to presenting SLiM criteria (1 with ≥60% bone marrow plasma cells, nine with a free light chain ratio ≥100), and one due to missing data (Supporting Information S1: Figure 2). The median age at diagnosis was 72 years (interquartile range [IQR]: 62–79) and 41% of patients were women. The median SMP size was 1.7 g/dL (IQR: 0.8–2.4), the median FLCr was 8.0 (3.2–23.0), and the median percentage of BMPC was 15% (IQR: 11.5–23.0). Similarly to other series of patients with SMM, IgG was the predominant heavy chain isotype (55.4%) and kappa the predominant light chain type (56.6%), as described in Table 1.
Table 1.
Baseline characteristics of patients at diagnosis.
| Characteristics | Patients n = 83 |
|---|---|
| Age at diagnosis, years (IQR) | 72 (62–79) |
| Gender | |
| Men/women, n (%) | 49 (59)/34 (41) |
| Hemoglobin (g/dL), median (IQR) | 13.5 (11.8–14.2) |
| Serum calcium (g/dL), median (IQR) | 9.3 (9–9.6) |
| Serum creatinine (mg/dL), median (IQR) | 0.79 (0.7–1.02) |
| Serum albumin (g/L), median (IQR) | 40.0 (38.0–42.0) |
| Β‐2‐microglobulin (mg/L), median (IQR) | 2.6 (2.1–3.5) |
| M‐protein (g/dL), median (IQR) | 1.7 (0.8–2.4) |
| % Bone marrow plasma cells, median (IQR) | 15.0 (11.5–23.0) |
| M‐protein type, n (%) | |
| IgG | 46 (55.4) |
| IgA | 31 (37.3) |
| Bence Jones | 3 (3.6) |
| Others | 3 (3.6) |
| Light chain type, n (%) | |
| Kappa | 47 (56.6) |
| Lambda | 35 (42.2) |
| Biclonal | 1 (1) |
| Free light chain ratio (IQR) | 8.0 (3.2–23.0) |
| Presence of an evolving pattern, n (%) | 21.0 (26.9) |
Abbreviations: Ig, immunoglobulin; IQR, interquartile range.
Among the 83 patients included in the study, 35 progressed during follow‐up (34 to symptomatic myeloma and one to systemic light‐chain AL amyloidosis) and five patients died without progression. Seven patients were censored: five at the time of inclusion in a clinical trial for high‐risk SMM, one due to progression during the first 6 months after diagnosis, and another due to loss of follow‐up before 1 year. The median follow‐up of the series was 4.6 years (IQR: 2.9–6.4) with the median TTP of 7.6 years (4.7‐not reached [NR]). The PR at 2 and 4 years was 19.8% (95% CI: 12.4–30.7) and 33.7% (95% CI: 23.5–46.6), respectively (Figure 1).
Figure 1.
Time to progression (TTP) to symptomatic multiple myeloma for all patients included.
Risk stratification models
The 2/20/20 IMWG and the PETHEMA score were used at diagnosis to classify the patients into groups according to the risk of progression to symptomatic MM. The cumulative incidence of progression for every risk group is shown in Figure 2.
Figure 2.
Progression rate to symptomatic multiple myeloma of every risk group according to the prognostic score used: (A) IMWG 2/20/20, (B) Evolving pattern.
Using the 2/20/20 IMWG risk stratification model, the 2‐year PR was 3.3% (95% CI: 0.5–21.4) in the low‐risk, 9.3% (95% CI: 2.4–32.4) in the intermediate‐risk, and 54% (95% CI: 34.9–74.9) in the high‐risk group of patients. The 4‐year PR was 11.2% (95% CI: 3.7–31.3) for the low‐risk, 20.9% (95% CI: 8.3–47.3) for the intermediate‐risk, and 82.7% (95% CI: 59.6–96.7) for the high‐risk group, as depicted in Figure 2A. The median TTP was not reached in the low‐risk group (95% CI: 7.57–NR), was 6.4 years (95% CI: 4.7–NR) in the intermediate‐risk group, and 1.6 years (95% CI: 1.0–NR) in the high‐risk group.
According to the PETHEMA risk model, the PR at 2 years for the low‐risk group was 5.6% (95% CI: 0.8–33.4), 17.4% (95% CI: 6.9–39.9) for the intermediate‐risk, and 27.9% (95% CI: 16.0–45.7) for the high‐risk group. The 4‐year PR was 5.6% (95% CI: 0.8–33.4) in the low‐risk group, 40.9% (95% CI: 22.4–66.5) in the intermediate‐risk, and 45.2% (95% CI: 28.6–65.9) in the high‐risk group (Supporting Information S1: Figure 3). The median TTP was not reached in the low‐risk (95% CI: 6.4–NR) and the intermediate‐risk (95% CI: 3.6–NR) groups, while it was 4.2 years (95% CI: 2.4–7.6) in the high‐risk group.
The evolving pattern
The evolving pattern was recognized in 26.9% of patients. The median time from SMM diagnosis to recognition of evolving type was 9 months (IQR: 6–12). The PR at 2 and 4 years was 48% (95% CI: 29.4–70.9) and 64.9% (95% CI: 44.2–84.7) for the evolving type, while it was 7.5% (95% CI: 2.9–18.7) and 20.0% (95% CI: 10.7–35.4) for the nonevolving type, respectively. The median time from recognition of an evolving pattern to progression was 11.6 months (IQR: 8.0–21.7). The median TTP for the group of patients with an evolving pattern was 2.4 years (95% CI: 1.3–NR) and was not reached in the nonevolving group (95% CI: 6.4–NR) as represented in Figure 2B.
New Risk Stratification Model (2/20/20‐Evolving)
After adding the evolving versus nonevolving classification to the 2/20/20 IMWG model, six curves were drawn, as represented in Figure 3A. These curves were merged into three new groups according to the 2‐year TTP and subhazard ratio to create a new prognostic classification system: low (IMWG low or intermediate risk and nonevolving pattern), intermediate (IMWG low or intermediate risk and evolving pattern, or high risk and nonevolving pattern), and high (IMWG high risk and evolving pattern), as shown in Figure 3B and Supporting Information S1: Table 1. The 2‐ and 4‐year PR in the new low‐risk group were 2.4% (95% CI: 0.3–15.7) and 5.0% (95% CI: 1.3–18.7), respectively. In the new intermediate group, the 2 and 4‐year PR were 20% (95% CI: 9.3–43.2) and 56.0% (95% CI: 36.4–77.7), respectively. Finally, in the new high‐risk group, the 2 and 4‐year PR were 88.9% (95% CI: 61.2–99.4) and 100%, respectively. Figure 3C represents in a Sankey diagram the flow of patients from the initial IMWG to the final risk group according to the evolving or nonevolving status.
Figure 3.
(A) Risk of progression to symptomatic multiple myeloma (MM) for each prognostic group according to 2/20/20 score, adding the presence or not of an evolving pattern to divide each risk group into two different curves. (B) Risk of progression to symptomatic MM for the three new groups defined by the new score. (C) Sankey diagram representing the change of risk group for each patient depending on the initial 2/20/20 classification and whether they present later on an evolving pattern or not. E, evolving; H, high risk; I, intermediate risk; L, low risk; NE, nonevolving; NH, new high risk; NI, new intermediate risk; NL, new low risk.
Performance measures for the New Risk Stratification Model
Regarding the performance of this novel score, the area under the curve (AUC) of the new score was greater than those of the IMWG and of the Evolving score independently. The 2‐year‐AUC PETHEMA was 0.65, the IMWG was 0.83, and 0.77 for Evolving, whereas the 2/20/20‐Evolving AUC was 0.89 (p = 0.02). The AUC for each score is represented in Supporting Information S1: Figure 4. In addition, the concordance of the new score measured with the C‐index was 0.93, higher than the PETHEMA (0.73), the 2/20/20 IMWG (0.86), and the evolving pattern (0.85) by themselves. The specificity and the positive predictive value for detecting high‐risk patients at 2 years with the new model were very high (98% and 88%, respectively) compared to the other risk stratification models (PETHEMA 59% and 27%; IMWG 83% and 52%; Evolving 83% and 49%, respectively) as shown in Table 2. Decision curve analysis indicated that the score was clinically useful, as shown in Figure 4. The new risk score detects most patients who will progress to symptomatic MM, although the net benefit is somewhat smaller, mainly because very few patients will be categorized as high risk. A great number of patients must be evaluated to find a patient at high risk of progression.
Table 2.
Performance of the new 2/20/20‐Evolving model.
| Model | Sensibility | Specificity | PPV | NPV | C‐index | 2 y AUC | 2y‐TTP (95% CI) | 4y‐TTP (95% CI) |
|---|---|---|---|---|---|---|---|---|
| Pethema | 64.7 | 58.6 | 26.6 | 87.7 | 0.73 | 0.65 | Low 5.56% (0.8–33.4) | Low 5.56 (0.8–33.4) |
| Int 17.4% (6.9–39.9) | Int 40.9 (22.4–66.5) | |||||||
| High 27.9 (16.0–45.7) | High 45.2 (28.6–65.9) | |||||||
| 2/20/20 | 78.3 | 82.8 | 51.3 | 94.3 | 0.86 | 0.83 | Low 3.3% (0.5–21.4) | Low 11.2 (3.7–31.3) |
| Int 9.3% (2.4–32.4) | Int 20.9 (8.3–47.3) | |||||||
| High 53.8% (34.9–74.9) | High 82.7 (59.6–96.7) | |||||||
| Evolving | 71.5 | 82.8 | 49.0 | 92.6 | 0.85 | 0.77 | NE 7.5% (2.9–18.7) | NE 20.0 (10.7–35.4) |
| E 48% (29.4–70.9) | E 64.9 (44.2–84.7) | |||||||
| 2/20/20 + Evolving | 56.8 | 98.3 | 88.4 | 90.7 | 0.93 | 0.89 | Low 2.4 (0.2–15.7) | Low 5.0 (1.3–18.7) |
| Int 20.9 (9.3–43.2) | Int 56.1 (36.4–77.7) | |||||||
| High 88.9 (61.2–99.4) | High 100 (100–100) |
Abbreviations: AUC, area under the curve; CI, confidence interval; E, Evolving; Int, intermediate; NE, nonevolving; NPV, negative predictive value; PPV, positive predictive value; TTP, time to progression; y, year.
Figure 4.
Decision curve analysis for the detection at 2 years of high‐risk patients.
DISCUSSION
The appearance of new treatment strategies, more effective and less toxic, has increased the interest in developing more accurate risk scores to identify those patients with SMM who will progress and those who will not. In this study, we initially classified 83 patients according to the 2/20/20 IMWG risk score and later reclassified them into three new categories according to the presence of an evolving versus non‐evolving pattern during follow‐up. The PR applying the 2/20/20 IMWG risk stratification model was 3.3% for low‐risk patients and 54% for high‐risk patients, a similar result to the previously published data. Using our new 2/20/20‐Evolving score, we identified a very high‐risk group with an 88.9% PR at 2 years, whereas the PR for the low‐risk subset of patients was only 2.4%. Thus, our score has a higher positive predictive value (88% vs. 49%), accurately identifying patients who will progress and could be eligible for treatment.
In 2014, the revised IMWG criteria identified an ultra‐high‐risk group of patients (defined by either BMPC >60%, FLCr >100, or >1 focal lesion on MRI) with an 80% PR at 2 years that were excluded from the SMM definition and considered candidates for treatment as symptomatic MM. 5 Our 2/20/20‐Evolving high‐risk group shows a similar PR (89%), supporting that these patients should also be considered for intensive follow up and treatment.
A recent nationwide study of MM precursors in Iceland found a 0.5% SMM in the total population and increased up to 1.1% in individuals over 70 years old, a higher incidence of SMM than what had been reported, indicating that it might not be as infrequent as thought. 14 SMM is a heterogeneous disease, which includes patients with an “MGUS‐like” illness that will not progress to MM and patients with a rapidly progressive disease. In a recent study, BMPC flow cytometry parameters were used to classify patients and identified a group of SMM (18%) with an “MGUS‐like” phenotype and a 4.5% risk of progression. 2 The use of risk scores in the management of SMM can help identify individuals at high risk for progression and guide the decision‐making process regarding the initiation of treatment. The current treatment approach for all patients is watchful waiting irrespective of the risk level, but several attempts to treat high‐risk patients have been made in the context of clinical trials. Randomized data have shown a progression‐free survival benefit in treating high‐risk SMM patients. 15 , 16 Due to this evidence, this approach is considered for high‐risk patients in some expert recommendations 17 to treat before the appearance of end‐organ damage, despite the lack of approval by the regulatory agencies. 18
Therefore, it is of vital importance to identify patients with high risk of progression who could benefit from early intervention. Updated risk stratification models, such as the 2/20/20 model, are used worldwide in clinical practice. This model has some limitations: it only assesses the initial tumor burden, without taking into account the decreasing risk of progression over the years and the biological tumor characteristics, closely related to the evolution over time of the tumor burden. 19 Approximately 50% of high‐risk patients did not progress within 2 years from the diagnosis, with the risk of overtreating patients. 20
Some groups have tried to re‐apply the risk scores over time to dynamically reclassify patients. This approach also faces limitations due to the requirement of bone marrow studies in some risk scores, including cytogenetic or molecular studies, and advanced body imaging. Moreover, if the risk of progression to MM decreases with time, re‐applying risk stratification scores postdiagnosis may not accurately stratify high‐risk SMM patients. 21 More recently, the PANGEA study developed a dynamic predictive model based on multiple risk factors, including monoclonal protein, FLC ratio, creatinine, hemoglobin trajectory, and age to calculate the risk of progression in clinical practice with a greater c‐index than the 2/20/20 model. 22
Other groups have developed new minimally invasive strategies to dynamically assess the risk of progression. In this sense, the assessment of circulating tumor cells might replace BMPC when assessing the risk of progression with the 2/20/20 IMWG score with comparable outcomes. However, this technology is not available in most centers and is still under investigation. 23
The sequential measuring of the SMP is a cheap, non‐invasive, widely available test that adds value to the existing and mostly used 2/20/20 IMWG classification system. A model based on the trajectory of biomarkers over time has also included an evolving hemoglobin and an evolving FLCr. 24 However, the lack of clear cutoffs to define some high‐risk biomarkers made some models difficult to apply in clinical practice.
Our study faces some limitations: all patients were seen at a tertiary care center specialized in plasma cell disorders, bringing a selection bias. This is a retrospective single‐center study with a limited number of patients, but quite similar to other previously published studies. 7 , 25 , 26 Furthermore, the retrospective nature of this study implies missing laboratory data and some patients did not have advanced imaging available. These results need to be validated in a large prospective international cohort of patients. Lack of availability of whole‐body MRI is common in many centers, an issue in identifying patients with the SLIM category. 6 The data regarding FLC over time was insufficient to define an “evolving” pattern, due to the lack of availability in our center for many years. In our study, only three Bence‐Jones SMM patients were included; however, in these patients, an “evolving” SMP pattern might not be useful in predicting the risk of progression. The definition of an “evolving” FLC pattern should be evaluated in a bigger cohort of patients. Genetic alterations allow us to identify high‐risk patients and are included in the extended IMWG 2/20/20 risk score; we could not include this data due to the limited cytogenetic information available in our series. However, although the 2/20/20 IMWG study included 1996 patients, less than a third of them had available cytogenetic data, 20 a limitation also faced in other similar studies. 6 , 22
In conclusion, the presence of an evolving pattern of SMM, characterized by the progressive increase of the SMP, is a well‐identified risk factor for progression to symptomatic MM. The sequential measurement of the SMP is non‐invasive and widely available. The dynamic assessment of the evolving pattern added to the IMWG classification (2/20/20‐Evolving) in patients with SMM, allows for reclassifying of the progression risk during the first year after diagnosis, resulting in a more precise stratification and identification of a subgroup of patients with a very high risk of progression who might be eligible for early treatment strategies.
AUTHOR CONTRIBUTIONS
Anna de Daniel, Luis Gerardo Rodríguez‐Lobato, and Carlos Fernández de Larrea designed the study. Luis Gerardo Rodríguez‐Lobato and Anna de Daniel analyzed the data, performed statistical analysis, and wrote and reviewed the manuscript. Luis Gerardo Rodríguez‐Lobato, Natalia Tovar, Anna de Daniel, Ignacio Isola, Carlos Fernández de Larrea, M. Teresa Cibeira, David F. Moreno, Aina Oliver‐Caldés, Laura Rosiñol, and Joan Bladé treated patients and reviewed the paper. All authors approved the final version of the manuscript.
CONFLICT OF INTEREST STATEMENT
Luis Gerardo Rodríguez‐Lobato: Honoraria and travel grants from Janssen, Amgen, GSK, and Sanofi; Aina Oliver‐Caldés: Travel grants from Janssen. David F. Moreno: Honoraria and travel grants from Janssen; M. Teresa Cibeira: Honoraria from Janssen, Sanofi, Amgen, Akcea; Laura Rosiñol: Honoraria from BMS, Celgene, Amgen, Takeda, Sanofi, Janssen, GSK; Joan Bladé: Honoraria from Janssen, Celgene, BMS, Amgen, Takeda, Oncopeptides; Carlos Fernández de Larrea: Consultancy: BeiGene, Sanofi, GSK, BMS, Janssen; Honoraria: BeiGene, Pfizer, Sanofi, GSK, Takeda, BMS, Janssen; Research Funding: Takeda, Amgen, BMS, Janssen. The rest of the authors declare no competing interests.
FUNDING
This work has been supported in part by grants from the Instituto de Salud Carlos III (ISCIII) and cofunded by the European Union (FIS PI22/00647 and ICI19/00025), 2021SGR01292 (AGAUR; Generalitat de Catalunya), and Asociación Española Contra el Cancer (AECC) LABAE21971FERN. LGRL has received funding from the CRIS Cancer Foundation (project number: PR_TMT_2023‐6).
Supporting information
Supporting information.
ACKNOWLEDGMENTS
The authors thank Esther Bladé for her technical support in this research.
Contributor Information
Luis Gerardo Rodríguez‐Lobato, Email: lgrodriguez@clinic.cat.
Carlos Fernández de Larrea, Email: cfernan1@clinic.cat.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1. Kyle RA, Child JA, Anderson K, et al. Criteria for the classification of monoclonal gammopathies, multiple myeloma and related disorders: a report of the International Myeloma Working Group. Br J Haematol. 2003;121(5):749‐757. 10.1046/j.1365-2141.2003.04355.x [DOI] [PubMed] [Google Scholar]
- 2. Burgos L, Tamariz‐Amador L‐E, Puig N, et al. Definition and clinical significance of the monoclonal gammopathy of undetermined significance–like phenotype in patients with monoclonal gammopathies. J Clin Oncol. 2023;41:3019‐3031. [DOI] [PubMed] [Google Scholar]
- 3. Rosiñol L, Bladé J, Esteve J, et al. Smoldering multiple myeloma: Natural history and recognition of an evolving type. Br J Haematol. 2003;123(4):631‐636. 10.1046/j.1365-2141.2003.04654.x [DOI] [PubMed] [Google Scholar]
- 4. Kyle RA, Remstein ED, Therneau TM, et al. Clinical course and prognosis of smoldering (asymptomatic) multiple myeloma. N Engl J Med. 2007;356:2582‐2590. 10.1056/NEJMoa070389 [DOI] [PubMed] [Google Scholar]
- 5. 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‐e548. 10.1016/S1470-2045(14)70442-5 [DOI] [PubMed] [Google Scholar]
- 6. Lakshman A, Rajkumar SV, Buadi FK, et al. Risk stratification of smoldering multiple myeloma incorporating revised IMWG diagnostic criteria. Blood Cancer J. 2018;8(6):59. 10.1038/s41408-018-0077-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Pérez‐Persona E, Vidriales M‐B, Mateo G, et al. New criteria to identify risk of progression in monoclonal gammopathy of uncertain significance and smoldering multiple myeloma based on multiparameter flow cytometry analysis of bone marrow plasma cells. Blood. 2007;110:2586‐2592. 10.1182/blood-2007-05-088443 [DOI] [PubMed] [Google Scholar]
- 8. Fernández De Larrea C, Isola I, Pereira A, et al. Evolving M‐protein pattern in patients with smoldering multiple myeloma: impact on early progression. Leukemia. 2018;32:1427‐1434. 10.1038/s41375-018-0013-4 [DOI] [PubMed] [Google Scholar]
- 9. Ravi P, Kumar S, Larsen JT, et al. Evolving changes in disease biomarkers and risk of early progression in smoldering multiple myeloma. Blood Cancer J. 2016;6:e454. 10.1038/bcj.2016.65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Delgado J, Pereira A, Villamor N, Lopez‐Guillermo A, Rozman C. Survival analysis in hematologic malignancies: recommendations for clinicians. Haematologica. 2014;99(9):1410‐1420. 10.3324/haematol.2013.100784 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Schultz LR, Peterson EL, Breslau N. Graphing survival curve estimates for time‐dependent covariates. Int J Methods Psychiatr Res. 2002;11(2):68‐74. 10.1002/mpr.124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Longato E, Vettoretti M, Di Camillo B. A practical perspective on the concordance index for the evaluation and selection of prognostic time‐to‐event models. J Biomed Inf. 2020;108:103496. 10.1016/j.jbi.2020.103496 [DOI] [PubMed] [Google Scholar]
- 13. Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565‐574. 10.1177/0272989X06295361 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Thorsteinsdóttir S, Gíslason GK, Aspelund T, et al. Prevalence of smoldering multiple myeloma based on nationwide screening. Nat Med. 2023;29(2):467‐472. 10.1038/s41591-022-02183-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Mateos M‐V, Hernández M‐T, Giraldo P, et al. Lenalidomide plus dexamethasone for high‐risk smoldering multiple myeloma. N Engl J Med. 2013;369:438‐447. 10.1056/NEJMoa1300439 [DOI] [PubMed] [Google Scholar]
- 16. Joseph NS, Dhodapkar MV, Lonial S. The role of early intervention in high‐risk smoldering myeloma. Am Soc Clin Oncol Educ Book. 2020;40:1‐9. 10.1200/EDBK_278915 [DOI] [PubMed] [Google Scholar]
- 17. Rajkumar SV, Kumar S, Lonial S, Mateos MV. Smoldering multiple myeloma current treatment algorithms. Blood Cancer J. 2022;12:129. 10.1038/s41408-022-00719-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. National Comprehensive Cancer Network . Multiple Myeloma (Version 3.2023). Accessed February 2, 2024. http://www.nccn.org/
- 19. Gran C, Luong V, Bruchfeld JB, et al. Dynamic follow‐up of smoldering multiple myeloma identifies a subset of patients at high risk of progression. Am J Hematol. 2021;96(3):E63‐E65. 10.1002/ajh.26062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Mateos MV, Kumar S, Dimopoulos MA, et al. International Myeloma Working Group risk stratification model for smoldering multiple myeloma (SMM). Blood Cancer J. 2020;10(10):102. 10.1038/s41408-020-00366-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Visram A, Rajkumar SV, Kapoor P, et al. Assessing the prognostic utility of smoldering multiple myeloma risk stratification scores applied serially postdiagnosis. Blood Cancer J. 2021;11(11):186. 10.1038/s41408-021-00569-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cowan A, Ferrari F, Freeman SS, et al. Personalised progression prediction in patients with monoclonal gammopathy of undetermined significance or smouldering multiple myeloma (PANGEA): a retrospective, multicohort study. Lancet Haematol. 2023;10(3):e203‐e212. 10.1016/S2352-3026(22)00386-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Termini R, Žihala D, Terpos E, et al. Circulating tumor and immune cells for minimally invasive risk stratification of smoldering multiple myeloma. Clin Cancer Res. 2022;28(21):4771‐4781. 10.1158/1078-0432.CCR-22-1594 [DOI] [PubMed] [Google Scholar]
- 24. Wu V, Moshier E, Leng S, et al. Risk stratification of smoldering multiple myeloma: predictive value of free light chains and group‐based trajectory modeling. Blood Adv. 2018;2(12):1470‐1479. 10.1182/bloodadvances.2018016998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Cocito F, Mangiacavalli S, Ferretti VV, et al. Smoldering multiple myeloma: the role of different scoring systems in identifying high‐risk patients in real‐life practice. Leuk Lymphoma. 2019;60(12):2968‐2974. 10.1080/10428194.2019.1620948 [DOI] [PubMed] [Google Scholar]
- 26. Piraud M, Wennmann M, Kintzelé L, et al. Towards quantitative imaging biomarkers of tumor dissemination: a multi‐scale parametric modeling of multiple myeloma. Med Image Anal. 2019;57:214‐225. 10.1016/j.media.2019.07.001 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.