MM: Multiple Myeloma; SEER: Surveillance, Epidemiology, and End Results; M-proteins: Monoclonal Immunoglobulin Proteins; CRAB: Hypercalcemia, Renal Damage, Anemia, and Bone Lesions; NDMM: Newly Diagnosed Multiple Myeloma; SMM: Smoldering Multiple Myeloma; MGUS: Monoclonal Gammopathy of Undetermined Significance; DL: Deciliter; IgM: Immunoglobulin M; IgG: Immunoglobulin G; IgA; Immunoglobulin A; IMWG: International Myeloma Working Group; PFS: Progression-free Survival; FLC: Free Light Chain; FISH: Fluorescence In situ Hybridization; IgM: Immunoglobulin M; MRI: Magnetic Resonance Imaging; CT: Computerized Tomography; PET-CT: Fludeoxyglucose (F18) Positron Emission Tomography with Computerized Tomography; IgD: Immunoglobulin D; IgE: Immunoglobulin E; AL: Immunoglobulin Light Chain Amyloidosis; AHL: Immunoglobulin Heavy and Light Chain Amyloidosis; AH: Immunoglobulin Heavy Chain Amyloidosis; ISS: International Staging System; LDH: Lactate Dehydrogenase; AHSCT: Autologous Hematopoietic Stem Cell Transplant; R-ISS: Revised International Staging System; OS: Overall Survival; Mo: Months; FU: Follow-up; IFM: Intergroupe Francophone du Myélome; R2-ISS: Second Revised International Staging System; NCCN: National Comprehensive Cancer Network; RDW: Red Blood Cell Distribution Width; MRD: Minimal/Measurable Residual Disease; CR: Complete Response; VGPR: Very Good Partial Response; ASO-PCR: Allele-specific Oligonucleotide Polymerase Chain Reaction; MFC: Multiparameter Flow Cytometry; NGS: Next-generation Sequencing; FDA: Food and Drug Administration; IStopMM: Iceland Screens, Treats, or Prevents Multiple Myeloma; PANGEA: Precursor Asymptomatic Neoplasms by Group Effort Analysis; MAPK: Mitogen-activated Protein Kinase; NF-kB: Nuclear Factor k-light Chain-enhancer of Activated B cells; APOBEC: Apolipoprotein B mRNA Editing Catalytic polypeptide-like; CoMMpass: Relating Clinical Outcomes in Multiple Myeloma to Personal Assessment of Genetic Profile; PROMMIS: Prospective Observational Multiple Myeloma Impact Study; IRMMa: Individualized Risk for Multiple Myeloma Model; DSS; Durie-Salmon Staging

Multiple myeloma (MM) is the second most common hematologic malignancy after lymphomas, affecting predominantly older adults with a median age of 69 years, based on the U.S. Surveillance, Epidemiology, and End Results (SEER) data [1]. MM is a neoplasm of plasma cells that infiltrate the bone marrow and may infiltrate extramedullary tissues, usually accompanied by increased levels of monoclonal immunoglobulin proteins (M-proteins), with the causation of end-organ damage, characterized by hypercalcemia, renal disease, anemia, or bone lesions (“CRAB” features) as the classically described clinical features [2-4]. In the USA, 35,780 new cases of MM, including 19,520 males and 16,260 females, and 12,540 deaths due to this disease, including in 7,020 males and 5,520 females are estimated to occur in 2024 [5]. The incidence per 100,000 U.S. persons of all ethnicities is higher in males at 8.7 than in females at 5.9, with the highest rates in non-Hispanic black males and females at 17.1 and 13.0, respectively [1]. In 2022, the global prevalence and incidence of MM were ~540,000 and 187,952, respectively, with 121,388 deaths [6]. The worldwide age-standardized incidence rates were the highest in the North American, Northern, Western, and Southern European, Australian, and New Zealand regions at 3.2-7.0 per 100,000 males and 2.5-6.2 per 100,000 females [6]. The global incidence is increasing and associated with a higher prevalence of physical inactivity, overweight, obesity, and diabetes [7]. In the last two decades, much has been learned about this unique hematologic malignancy, with progressive, continuous, and ongoing advances in patient survival. Still, there are unmet needs, and a cure is not yet apparent. The purpose of this review is to briefly describe significant advances in MM that have led to improvements in survival, including a brief description of the current diagnostic criteria of MM and its precursor conditions, staging, and prognosis of newly diagnosed MM (NDMM), treatment response criteria, including minimal or measurable residual disease (MRD) assessment, progress in the evaluation of the risk of progression of precursor states to active MM, and recent advances in the genomics of NDMM and smoldering multiple myeloma (SMM), which still need to be used routinely for determining prognostic risk in NDMM and progression risk in precursor states to enable precise identification of all patients for individualized risk-adapted clinical management and treatment regimens.

Multiple myeloma consistently evolves from a precursor condition, monoclonal gammopathy of unknown significance (MGUS) [8], but the converse is not true. MGUS, as defined by the presence of fewer than 3 grams/deciliter (dL) of serum M-proteins, absent urinary M-proteins, or moderate amounts of light chains in the urine, the absence of any end-organ damage, and fewer than 10% plasma cells in the bone marrow [9], progresses to MM, Waldenstrom’s macroglobulinemia or light chain amyloidosis at the rate of 1% per year, notably, with a greater risk of death due to causes other than progression to plasma cell neoplasms [2,9,10]. Two types of MGUS, non-immunoglobulin M (non-IgM) MGUS, and light chain MGUS are caused by monoclonal plasma cells, and these types of MGUS are precursors for MM [8]. In contrast, IgM MGUS is caused by lymphoplasmacytoid cells, and this type of MGUS may progress to Waldenstrom’s macroglobulinemia or immunoglobulin-related amyloidosis [11].

The second precursor neoplasm, SMM, was defined by the International Myeloma Working Group (IMWG) in 2003 [12] by the presence of ≥3% serum M-proteins, or ≥10% plasma cells in the bone marrow, in the absence of any end-organ damage [12]. In 2007, a study of 276 patients with SMM showed the following risks of progression to MM with end-organ damage or amyloidosis: 10% per year for the first 5 years, 3% per year for the next 5 years, and 1% per year for the subsequent 10 years, with a cumulative 73% risk of progression at 15 years [13,14]. Of these 276 patients, only six (2%) had bone marrow plasmacytosis of >60%, and these six patients had a median progression-free survival (PFS) of 7.7 months, with disease progression or death in 83% [15]. A free light chain (FLC) ratio of >8 or <0.125 in the serum of patients with SMM showed a >50% risk of progression to active disease [16]. The IMWG consensus guidelines for monitoring MGUS and SMM were published in 2010 [17].

A similarly high risk of progression was shown to be present in small percentages of patients having ≥60% bone marrow plasma cells in three additional cohorts of SMM during 2011-2013, with 95% of those high-risk patients progressing to active MM within 2 years: (1) 3.2% (n = 21) of 655 patients with a median time to progression of 7 months [15]; (2) 12.5% (n = 12) of 96 patients, also with a highly abnormal FLC ratio of ≥100 or ≤1/100 in the serum, progressing in less than 18 months [18]; and (3) in 4% (n= 6) of 135 patients in the third cohort [19]. In another study, an FLC ratio of >100 in the serum of 90 patients with SMM showed a 79% risk of progression to active MM or light-chain amyloidosis [20].

Further, in 2007, the Spanish group showed a high proportion (95%) of immunophenotypically aberrant plasma cells among total bone marrow plasma cells by multiparametric flow cytometric immunophenotypic analysis to have a significant risk of progression in both precursor states, MGUS and SMM [21]. In conjunction with a reduction from normal levels of one or two uninvolved immunoglobulins (termed immunoparesis), the presence of 95% aberrant plasma cells based on the analysis of CD38, CD138, CD19, CD45, and CD56 expression, showed a 72% risk of progression of SMM to active MM [21]. Based on these features, the Spanish Myeloma group showed delayed progression of high-risk SMM by early treatment with lenalidomide, and dexamethasone compared with observation in a randomized phase 3 clinical trial published in 2013 [22]. In another study investigating radiologic imaging, the presence of >1 focal lesion detected by whole-body magnetic resonance imaging in 149 patients with SMM predicted disease progression to active MM [23].

Genetically, the neoplastic cells in MM are considered to harbor either chromosomal hyperdiploidy or chromosomal translocations involving the immunoglobulin heavy chain gene locus at 14q32 [24]. These abnormalities are considered primary genetic events that are also present in the earlier stage of MGUS. They are insufficient to cause MM, which requires additional secondary genetic abnormalities such as copy number changes, DNA hypomethylation, and acquired mutations to occur in the plasma cell (a post-germinal center mature B-cell) that had acquired a primary genetic abnormality while transiting the follicular germinal center or had inherited genetic variation [24]. The neoplastic plasma cells in MM are thought to derive from self-renewing myeloma-propagating cells due to interactions in the bone marrow niche causing the transformation of plasma cells to neoplastic myeloma cells [24].

The current diagnostic criteria for MM and its precursor neoplasms were developed by the IMWG in 2014 [11], as an update from the previous 2003 [12] and 2010 [17] criteria, to include very high-risk SMM as a type of active MM to enable treatment of those patients [11,25]. In addition to 10% clonal bone marrow plasma cells or a biopsy-proven bony or extramedullary plasmacytoma, the 2014 IMWG diagnostic criteria of MM required any one or more of myeloma-defining events, which then started to include any of the following additional biomarkers, referred to as “SLiM” in addition to the above-mentioned CRAB features, for “SLiM-CRAB” criteria [11]:

(1) Bone marrow plasmacytosis ≥60% (“S” for “sixty” in SLiM),

(2) Serum FLC involved: uninvolved ratio ≥100 (“Li” for “light chain” in SLiM), and

(3) More than one focal lesion on MRI studies, with each focal lesion being >5mm in size (“M” in SLiM) [11].

The 2014 IMWG group defined SMM to require the following two criteria [11]: (1) serum M-protein (IgG or IgA) ≥30 g/L or urinary M-protein ≥500 mg per 24 hours; or clonal bone marrow plasma cells 10% to 60%; or both; and (2) the absence of myeloma-defining events or amyloidosis [11].

SMM was further characterized for risk of progression to active MM by a retrospective single-institution Mayo Clinic study that identified 400 patients based on the 2014 IMWG criteria [26]. In that study, three readily determined variables, i.e., bone marrow plasma cells >20%, serum M-protein >2g/dL, and an abnormal serum FLC ratio >20, were considered to predict a 50% risk of progression at 2 years, if two or more factors were present [26]. In 2020, the IMWG confirmed this risk stratification model often termed “20/2/20” in an international retrospective study of 1,996 patients with SMM from 75 centers in 23 countries [27]. This risk model uses readily available clinical values to estimate tumor burden and can be applied in low-resource settings. Findings identified by fluorescence in situ hybridization (FISH), present in only a subset of their studied patients, could improve the risk model, and the IMWG authors noted that prospective studies would be needed to determine the clinical validity of incorporating cytogenetics and genomics data in identifying more precise risk of progression [27].

Figure 1 depicts the differential diagnosis of non-IgM MGUS and SMM, which are precursors of active MM, with active MM, which must be treated. The presence of any of the CRAB features of end-organ damage or myeloma-defining biomarkers (often termed SLiM-CRAB as explained above) must be absent in MGUS and SMM. The presence of any of the CRAB or SLiM-CRAB features, in conjunction with either 10% or greater clonal plasma cells in the bone marrow or a biopsy-proven plasmacytoma or both indicates a diagnosis of active MM, which requires treatment. The IMWG diagnostic criteria are used universally and have been included in the World Health Organization diagnostic pathology classification [28] and the parallel International Consensus classification of hematolymphoid neoplasms [29]. Since plasma cells are post-germinal center B cells, multiple myeloma is also termed plasma cell myeloma and is classified among lymphoid neoplasms arising from B cells [28,29].

Figure 2 shows the defining criteria for the diagnosis of the three types of MGUS, based on IMWG criteria [11]. The evaluation of the risks of progression of non-IGM MGUS and light chain MGUS to active disease is described in a subsequent section.

Figure 3 shows the defining criteria for solitary plasmacytoma with and without minimal bone marrow involvement.

Plasma cell leukemia is currently defined by the presence of 5% or greater circulating plasma cells in the peripheral blood smear of a patient diagnosed with multiple myeloma, according to an IMWG consensus in 2021 [31]. The earlier diagnostic criteria for this leukemia required the presence of >20% circulating plasma cells and an absolute circulating plasma cell count of 2 x 10⁹/L [32]. Plasma cell leukemia may be primary, i.e., present at the time of MM diagnosis, or secondary, i.e., present in relapsed or refractory MM in a patient with a previous diagnosis of MM [31]. Patients with MM having increased circulating plasma cells are expected to have poor overall survival and progression-free survival [33].

Refractory and relapsed multiple myeloma were defined by the IMWG [34] to design clinical trials using novel drugs [35]. According to the IMWG definitions, refractory myeloma refers to the refractoriness of the disease to treatment, which can occur after primary (or initial) treatment or after salvage therapy. Refractory myeloma is identified by non-responsiveness, i.e., failure to achieve a minimal response, or by progressive disease while on therapy or within 60 days of the last therapy [34]. Refractory myeloma may be “relapsed and refractory,” which occurs after relapse during salvage therapy, or it may be “primary refractory myeloma,” in which case the patient does not ever achieve a minimal response or better with any therapy [34].

In contrast, relapsed (or recurrent) myeloma is defined as myeloma that progresses after being previously treated but it does not meet the criteria for relapsed and refractory myeloma or primary refractory myeloma described above [34]. These IMWG definitions are used for the current treatments [36] and clinical trials (reviewed in [37]) for patients with relapsed or refractory MM.

An accurate assessment of prognostic risk at diagnosis of disease is critically important for patients and their treating clinicians. Multiple myeloma is a very heterogeneous disease, and clinical outcomes are variable with some patients considered as high-risk having worse outcomes than others in the same risk group [38]. The improved survival rates have been seen primarily in MM considered to have standard risk [39]. Even some patients considered as standard-risk MM may show early relapses or refractory disease or inferior outcomes compared to other patients with standard-risk MM, indicating that there is heterogeneity in the risk within the defined risk categories [39-41]. The term “functional high-risk myeloma” has been used for those patients who do not fall into a high-risk group at diagnosis by the current criteria but are recognized as high-risk after failure of front-line therapy, including relapse within 18 months of treatment initiation or 12 months of frontline autologous stem cell transplantation [42].

Historically, a clinical staging system for MM was first developed by Durie and Salmon in 1975, based on four variables: hemoglobin, serum calcium level, bone lesions on X-ray, and serum and urine M-protein levels, which provided three risk groups based on the tumor (myeloma) cell mass [43], shown as stages I-III in (Table 1). In 1990, serum b2-microglobulin was identified as the most powerful predictor of prognosis in MM [44]. This simple laboratory value, representing a disease factor, in conjunction with the serum albumin level (a patient factor), comprised the two variables required by the International Staging System (ISS) in 2005 to separate three prognostic risk groups [45], also shown in (Table 1). A subsequent comparison of the two staging systems in 2009 showed that the ISS was not prognostically reliable in patients treated by transplantation indicating that additional variables were required for prognostic risk stratification [46].

In 2010, a German study showed the independent prognostic significance of cytogenetic abnormalities in MM patients who underwent autologous stem cell transplantation [47]. In 2011, an IMWG consensus panel considered the following abnormalities to be high-risk in MM: high serum b2 microglobulin, low serum albumin, and chromosomal 13 or 13q deletion, t(4;14), and del(17p), if detected by chromosomal banding analysis, and t(4;14), t(14;16), and del(17p) if detected by FISH [48].

In 2012, the UK Medical Research Council group identified the following cytogenetic abnormalities to have adverse prognostic risk independent of the ISS: +1q21, del(17p13) and immunoglobulin heavy chain gene translocations t(4;14), t(14;16) and t(14;20) [49]. They defined risk groups based on the number of adverse cytogenetics abnormalities: (1) no adverse cytogenetics as favorable, (2) one adverse genetic lesion as intermediate, and (3) >1 adverse lesion as high-risk [49]. They also identified an ultra-high-risk group by combining >1 adverse genetics with the ISS stage II or III, thus improving risk stratification [49]. A combined genetic-ISS model, including patient age, improved prognostic risk stratification in an international collaborative IMWG study in 2013 [50]. Based on these previous studies [47,49,50], the IMWG consensus on risk stratification was published in 2014 [51]. The Mayo Clinic guidelines [52] for risk stratification, published in 2013 [52], were based primarily on cytogenetics findings, as follows: (1) high-risk: del(17p), t(14;16), and t(14;20) by FISH, (2) intermediate-risk: t(4;14) by FISH, cytogenetic deletion 13, hypodiploidy, and (3) standard-risk: all other abnormalities, including t(11;14) and t(6;14) by FISH [52]. A high-risk gene expression profiling signature was included in the Mayo Clinic guidelines for high-risk and plasma cell labeling index ≥3% for intermediate-risk groups, but the guidelines stated that these tests would be unavailable to most clinicians and were not recommended [52]. A 92-gene expression profiling signature was developed in 2012 for use in clinical trials; the authors emphasized that this signature had not been developed to evaluate risk in a single individual patient and was meant to be used in groups of at least 25 patients [53].

Subsequently, three independent variables were found to identify a risk of death due to early progression in patients treated with an autologous hematopoietic stem cell transplant (AHSCT): elevated serum lactate dehydrogenase (LDH), ISS stage 3, and adverse cytogenetics t(4;14), del(17p), or both genetic abnormalities [54]. The above-described laboratory values and cytogenetic abnormalities were evaluated in patient-level data and incorporated into the Revised ISS (R-ISS) in 2015 [55]. As shown in Table 1, the R-ISS created three stages I, II, and III, defined by the ISS stage, serum LDH, and the presence of high-risk or standard-risk cytogenetic abnormalities by FISH. However, the R-ISS identified only 9.5% of patients in stage III (highest risk in the system) with the maximum percentage being 61.9% for the intermediate stage II, for whom there were no inclusive definitions, except not being stratified as ISS stages I or III [55]. Based on the R-ISS, the IMWG guidelines for treating high-risk MM were published in 2016 [56].

In 2019, the Intergroupe Francophone du Myélome (IFM) developed and validated a prognostic index based on the clinical data and results for seven specific cytogenetic abnormalities, (17p), t(4;14), del(1p32), 1q21 gain, trisomy 3, trisomy 5 (a good prognostic factor), and trisomy 21 in 1,635 patients with NDMM enrolled in four clinical trials between 2000 and 2012 [57]. Bone marrow plasma cells were selected by CD138+ magnetic cell sorting, and only those samples with ≥70% plasma cells post-sorting were selected for cytogenetic analysis. FISH was performed for t(4;14)(p16;q32), and the other abnormalities were detected by a copy number and single nucleotide polymorphism array platform. If del(17p) was detected, the percentage of involved plasma cells was determined by an additional TP53/CEP17 FISH probe analysis, and del(17p) was included only if present in ≥60% of plasma cells [57,58]. Cytogenetics findings alone provided a better separation than the ISS of patients who died from those who survived and were better than the R-ISS in an external validation dataset in correlating the prognosis with the outcome. Three risk groups were identified: low, intermediate, and high. Low risk was present only if del(17p) and t(4;14) were both absent [57]. Less than 35% of patients with t(4;14) but most patients with del(17p) were identified as high-risk [57]. The hazard ratios for death were 6 to 15 times higher in high-risk compared to low-risk patients [57]. The t(14;16) was not included in their analysis since there have been studies showing that the presence of this abnormality does not confer a higher risk (reviewed in [57]); in their study, the risk did not differ due to the presence or absence of the t(14;16) abnormality [57].

Chromosome 1q abnormalities have been studied in MM for at least two decades. Amplification of 1q, representing ≥4 or more copies of 1q, and gain of 1q, representing three copies of 1q (two normal copies and one extra copy) have been shown to have a poor prognosis in MM, with the evidence being stronger for amplification of 1q than for gain of 1q (reviewed in [59]). The Second Revised ISS (R2-ISS) published in 2022 was based on 10,483 patients with NDMM enrolled in 16 international, multi-center clinical trials from 2005 to 2016, with a median follow-up of 75.5 months [60]. Patient-level data were examined in this study for factors required for the R-ISS and gain or amplification of 1q, referred to as 1q+ since separated data for gain or amplification of 1q were unavailable for these patients [60]. The factors that most affected the overall survival (OS) and PFS were the ISS stage, serum LDH, del(17p), t(4;14), and 1q+ [60]. These variables were used in the training set of 2,226 patients to derive an additive risk score, which determined four risk groups shown in Tables 2A and 2B [60]. The intermediate-risk group in the three-stage R-ISS was re-defined by the R2-ISS by defining four risk groups [60].

The biggest strength of this system is that the risk factors are possible to evaluate in most patients, requiring only simple laboratory tests and FISH studies, which are readily available in many institutions worldwide [60]. However, it does not yet include other poor significant prognostic factors as discussed by the authors [60], including the presence of circulating plasma cells [61-66], TP53 mutations, and extramedullary disease (defined in [67]), and patient factors such as frailty, which could be added to further identify high-risk in NDMM. Red blood cell distribution width (RDW) has been reported as a parameter that is increased in IgA MM and increases with progressive clinical stages of MM, and this value could be incorporated into prognostic models for MM [68]. The risk factors recognized as high-risk in NDMM and relapsed MM by the current U.S. National Comprehensive Cancer Network (NCCN) guidelines are shown in Figure 4 [69].

The first uniform criteria for treatment responses in MM were developed by the European Group for Blood and Bone Marrow Transplant in 1998 [70]. In 2006, the IMWG updated these criteria due to the need to compare precisely the effects of new therapies in clinical trials and use in the clinic for individual patients, with the following definitions [71]:

(1) “Complete response” (CR) required <5% bone marrow plasma cells, negative immunofixation in serum and urine, and disappearance of any soft tissue plasmacytomas.

(2) “Stringent CR” required CR plus normal serum FLC ratio and the absence of clonal cells in bone marrow by immunohistochemistry or immunofluorescence to evaluate the k: l ratio (with an abnormal k: l ratio being >4:1 or <1:2 based on ≥100 plasma cells).

(3) “Very good partial response” (VGPR) with serum and urine M-protein detectable by immunofixation but not by electrophoresis or ≥90% reduction in serum M-protein plus urine M-protein <100 mg per 24 hours; and

(4) Partial response with 50% reduction of serum M-protein and reduction in 24-hour urinary M-protein by 90% or <200 mg per 24 hours.

A serum FLC assay was added to evaluate treatment responses in patients with non-secretory and oligo-secretory disease [71].

In 2006 [71], “measurable disease” was defined by serum and urine M-protein and serum FLC levels, requiring the measurement of any one or more of the following three parameters: (1) serum M-protein ≥1 g/dL (≥10 gm/L), (2) urine M-protein ≥200 mg/24 hours, and (3) involved serum free light level ≥10 mg/dL (≥100 mg/L) with an abnormal serum FLC ratio [71]. In 2011, two-to-four-color flow cytometry was added to the response criteria as a technique for determining the k: l ratio for stringent CR; “immunophenotypic CR” was defined as stringent CR plus the absence of phenotypically aberrant (clonal) plasma cells in the bone marrow with a minimum of 1 million total bone marrow cells analyzed by multiparametric flow cytometry with > 4 colors, and “molecular CR” was defined as CR plus negative allele-specific oligonucleotide polymerase chain reaction (ASO-PCR) with a detection sensitivity of 1 in 10⁵ cells [34].

A decade ago, the treatments using newer therapies for multiple myeloma led to an urgent need to measure deeper responses to treatment that could detect residual disease with higher sensitivities than the traditional methods of measuring serum and urine M-proteins. It has been known for more than two decades that flow cytometric immunophenotypic analysis can detect normal plasma cells from neoplastic plasma cells in the same specimen [72,73]. Multiple publications in 2015 by international experts addressed MRD evaluation by multiparameter flow cytometry (MFC) in MM, including the need for a harmonized approach [74], data acquisition [75], data analysis and reporting [76], and quality control [77]. The IMWG consensus criteria for treatment response, published in 2016 [78], added responses to treatment after detecting MRD in bone marrow specimens by high sensitivity MFC (termed next-generation flow) or by next-generation sequencing (NGS) of the immunoglobulin heavy chain gene variable-diversity-junctional (VDJ) region at a minimum sensitivity level of 1 in 10⁵ bone marrow cells, and to detect extramedullary disease by sensitive radiologic imaging techniques, which form the current guidelines for MRD assessment in multiple myeloma [69,78].

As per guidelines from the U.S. Food and Drug Administration (FDA), the detection sensitivity of an MRD assay in hematologic malignancies is required to be 10-fold below the clinical decision-making threshold, which means that an assay’s analytical sensitivity of detecting 1 abnormal cell in 10⁶ cells is required for the clinical decision-making threshold of 1 in 10^-5 [79]. Comparing MFC with NGS, both methods can achieve the high sensitivity required for MRD evaluation in multiple myeloma [80-82]. MFC has a significant advantage of an in-built evaluation of the quality of the bone marrow sample being evaluated, since normally occurring bone marrow cells such as B cell precursors (hematogones), nucleated red blood cells, and mast cells can be identified by flow cytometry [76]. Also, normal and neoplastic plasma cells in the bone marrow can be distinguished by MFC, allowing a precise determination of MRD [83]. Although MFC requires a fresh sample with viable cells, the technique has been used worldwide even in developing countries with limited resources, and a diagnostic sample is not necessary [83,84]. In contrast, although NGS can be performed on an archived sample, the technique for MRD evaluation mandates the availability of a quality diagnostic bone marrow sample, ideally less than 48 hours old [74,84] to first establish the diagnostic clone, and high-quality DNA and high total DNA yield are critically required for MRD evaluation by NGS [81]. In addition, if the sample is transported to a reference laboratory for NGS, the cost increases and hemodilution in the sample analyzed for MRD may not be possible to evaluate for. Finally, the choice of method would depend on the instrumentation and expertise available at the institution [82].

Regardless of the selected method, both NGS and MFC depend on a bone marrow biopsy, which is well-known to pathologists for having patchy involvement by neoplastic plasma cells in MM. Even if the first pull in a bone marrow biopsy is dedicated for MRD analysis by MFC or NGS, that specimen might still not accurately represent the amount of disease even in the bone marrow, and another method must still be used for extramedullary disease assessment. Therefore, for MRD assays to be truly representative of the residual quantity of myeloma disease in the body, including involved extra-medullary and medullary foci, other methods using peripheral blood draws such as analysis of genomic variants in circulating tumor DNA [85] and circulating tumor cells [61] should be seriously considered for further development to be used on a wide scale to evaluate MRD in patients with multiple myeloma.

The 5-year survival rates for all patients with MM in the USA improved from 25% in 1975-1977 and 32% in 1995-1997 to 60% in 2013-2019 [5]. The increased survival in the last two decades is primarily due to improved therapies and novel agents. Thalidomide was introduced in clinical trials in MM in the late 1990s and lenalidomide and bortezomib in the early 2000s [86]. A registry-based Swedish study showed a 29% reduction in mortality in younger patients aged <70 years diagnosed in the 1994-2003 period [87]. Subsequently, a patient-level study showed that among patients treated from 1950 to 2005 in Sweden, overall survival improved the most in the 2000-2005 period [88]. This improvement was attributed to the use of high-dose melphalan before autologous hematopoietic stem cell transplantation and novel agents such as thalidomide [87,88].

Improvements in survival in the USA also occurred mainly in younger patients, based on SEER data, with patients aged <50 years showing 5- and 10-year relative survival of 56.7% and 41.3% in 2002-2004 and patients aged 50 to 59 years showing 5- and 10-year relative survival of 48.2% and 28.6% in 2000-2004, but only moderate improvement in ages 60 to 69 years, and no improvement in older patients [89]. In a Mayo Clinic cohort, OS was shown to be significantly improved in patients with MM diagnosed for the first time or with relapse after 2000 and treated with one or more of the novel agents, thalidomide, lenalidomide, bortezomib, compared with patients diagnosed in the 1990s [90]. In 2014, 6-year OS was shown to be improved from 31% to 56% primarily in older patients aged >65 years from 2001-2005 to the 2006-2010 groups of patients at the Mayo Clinic, a specialized institution in the USA [91]. The OS for Spanish patients with MM improved from 1980 to 2020, with median values of 22.4 months in 1980-1990, 37.4 months in 1991-2000, 61.8 months in 2001-2010, and 103.6 months in 2011-2020, attributed mainly to the use of novel agents in frontline therapy [92]. Of note, survival has also improved significantly in the 70-79 and >75 years age groups, as shown in large population-based studies from Germany [93], Norway [94], Sweden, and Denmark [95].

During 2007-2014, there were 128 publications for clinical trials conducted only in the USA for MM [96], reflecting the active clinical research in treating this disease, albeit with the under-representation of the patient populations with greater disease prevalence in the USA [96]. In 2012, lenalidomide as maintenance following autologous hematopoietic stem cell transplantation showed significantly improved progression-free survival (PFS) [97,98] and OS [97]. Also in 2012, carfilzomib was used for frontline treatment in combination with lenalidomide and dexamethasone in patients aged 35-81 years (with 43% >65 years of age) to show exceptional responses [99]. Bortezomib, combined with thalidomide and dexamethasone as induction therapy before, and consolidation therapy after, double AHSCTs for NDMM showed a superiority of the triplet regimen including a proteasome inhibitor and an immunomodulatory agent compared with thalidomide and lenalidomide alone [100,101], also confirmed after long-term follow-up [102].

Also in 2014, low-dose lenalidomide and dexamethasone until disease progression in a clinical trial of 1,623 patients with transplant-ineligible MM showed significantly improved PFS, and improved OS, compared with the standard-of-care combination of melphalan, prednisone, and thalidomide [103]. Updated results in 2016 confirmed the OS benefit of low-dose lenalidomide-dexamethasone, including a 14-month improvement in median OS in patients >75 years [104]. Daratumumab, an anti-CD38 human IGGk monoclonal antibody, that kills myeloma cells by antibody-dependent and complement-dependent cytotoxicity, was given breakthrough status [105] and first given accelerated approval in the USA in November 2015 to treat relapsed or refractory MM with ≥ three prior lines of therapy, including a proteasome inhibitor and an immunomodulatory agent (reviewed in [106]) or if refractory to both classes of drugs [107]. In 2016, unprecedented improvements in PFS were shown [108] in patients with relapsed or refractory MM by adding daratumumab to the combination of bortezomib and dexamethasone [109] or lenalidomide and dexamethasone [110]. The SWOG S0777 trial, which started in 2007, showed in 2017 a landmark improvement in OS to 75 months by adding bortezomib to lenalidomide and dexamethasone in NDMM [111,112].

The U.S. NCCN guidelines from April 2024 recommend triplet drug regimens as preferred combinations for frontline therapy in transplant and non-transplant candidates, or a quadruplet drug regimen also including an anti-CD38 agent such as daratumumab in transplant-eligible patients with MM [69]. The newer drug classes include proteasome inhibitors such as bortezomib, carfilzomib, and ixazomib (reviewed in [113,114]), immunomodulatory agents such as thalidomide [115,116], lenalidomide, and pomalidomide (reviewed in [106,116]), often combined with dexamethasone in triplet drug regimens. Despite the advances in therapies with improved survival rates, MM remains incurable, and the disease is characterized by several relapses [117]. The best chances of a sustained deep therapeutic response are in the frontline treatment of MM since the proportion of patients receiving therapy decreases with each successive line of therapy [118].

In the last decade, the sustained absence of minimal or preferably, measurable residual disease (MRD) at the minimal sensitivity level of detecting one abnormal plasma cell in 100,000 (or 10⁵) cells has emerged as an indicator of favorable prognosis in patients with MM [119]. Flow cytometry is the most readily available method worldwide with a fast turnaround time compared to next-generation sequencing to detect MRD [120] with ongoing efforts to harmonize high-sensitivity flow cytometry data analysis internationally to achieve conformity in clinical trials [83]. According to the NCCN guidelines from April 2024, MRD should be evaluated only with a suspected complete response, with MRD evaluation recommended after each stage of treatment [69]. Intense clinical research in the form of clinical trials using different treatment modalities, including anti-CD38 antibodies, chimeric antigen receptor T-cell therapy, and bispecific antibodies, has been conducted by the myeloma community toward curing the disease. Excellent treatment responses including high rates of undetectable MRD (considered as MRD-negativity) have been seen in transplant-eligible NDMM in the phase 3 PERSEUS (daratumumab, bortezomib (V), lenalidomide (R), and dexamethasone (D) vs. VRD) [121] and phase 2 GRIFFIN (Daratumumab-VRD vs. VRD) [122] trials, transplant-ineligible NDMM in the phase 3 MAIA (Daratumumab-RD vs. RD), [123] and phase 3 IMROZ (Isatuximab-VRD vs. VRD) [124]) trials, and relapsed or refractory MM [125-131]. The international MM community has been working toward the use of MRD as an endpoint as a surrogate for PFS in clinical trials for patients with MM [119,132-134] since 2009 (discussed in [134]). In April 2024, an oncologic advisory committee of the FDA approved MRD as an endpoint for accelerated approvals in clinical trials for transplant-eligible and transplant-ineligible NDMM and relapsed or refractory MM [135]. This approval is a significant milestone since it will allow a readout of clinical trials earlier than a clinical survival endpoint, and in the future, after being studied at the patient level, MRD evaluation could allow clinical decision-making during the treatment of MM in individual patients.

Despite the advances in novel therapies available to patients with MM, several barriers preclude access to novel agents, including AHSCT for treating patients with MM from ethnic minority populations in the USA, which many clinicians have studied in the USA [136-143]. A recent National Cancer Database study of 111,799 patients with MM diagnosed during 2004-2013 and who received AHSCT as primary treatment, comprising 77.5% non-Hispanic White, 15.1% non-Hispanic Black, 5.2% Hispanic, and 2% non-Hispanic Asian patients, identified socioeconomic factors that affected access to treatment [136]. Of note, insurance payer status significantly affected the access to an AHSCT for all ethnic groups [136]. In this context, the complex healthcare structure in the USA is described as an important structural barrier to accessing novel treatments [138,144]. In the entire cohort, only 10.6% of all non-Hispanic Black patients received an AHSCT [136]. In another 2008-2014 cohort of 28,450 patients undergoing AHSCT for MM, it was notable that post-transplant outcomes were similar, but there were differences in utilization rates of AHSCT in different ethnic groups [137]. The lowest rate of increase in AHSCT was seen in Hispanic, followed by non-Hispanic Black patients, compared with non-Hispanic White patients [137]. Hispanic and non-Hispanic Black patients are often younger than 60 years, compared to the median age of 69 years for MM, are likely to have advanced-stage disease, and are referred for transplant later than non-Hispanic White patients [137]. Access to novel therapies for MM is delayed in minority populations in the USA, as shown in several studies based on SEER data [139-141]. Globally, factors adversely affecting access to treatment include older age, female sex, and Black, Hispanic, or Asian ethnicity [145].

It is important to note that the above-described risks of progression from asymptomatic MGUS or SMM to active MM represent an average risk for patient populations. These risks do not apply to the risk of progression in individual patients, as described in an example in 2021 [146]. Also, tumor burden alone cannot predict whether the disease will progress or how fast the disease will progress. A longitudinal follow-up study in 2019 showed that the risk of progression in MGUS can change quickly from low or intermediate risk to high risk [147]. Of note, among the patients who progressed from MGS to MM, the majority (79%) progressed directly to MM without an intermediate SMM phase [147]. The adverse prognostic indicators for the progression of MGUS included an IgA isotype of the M-protein, ≥15g/L M-protein concentration, serum FLC (k:l) ratio of <0.1 or >10, and uninvolved immunoglobulins below the level of normal; adding points based on these variables can stratify progression risk of MGUS into low, intermediate, and high-risk categories [146,147]. Figure 5 shows these risk factors and low-, intermediate-, and high-risk categories for progression of non-IgM MGUS and light chain MGUS [146,147].

In 2024, the Iceland Screens, Treats, or Prevents Multiple Myeloma (iStopMM), a prospective population-based study developed a multivariable prediction model in 1,043 individuals ≥40 years of age with MGUS to predict the risk of SMM, which provided greater benefit than the Mayo Clinic model to decide when to defer a bone marrow biopsy [148]. In the iStopMM study, individuals with SMM had a mean M-protein concentration of 0.62 (range 0.01-3.5) g/dL, and in 73% of persons with SMM, there were 11-20% bone marrow plasma cells [149].

Personalized risks of progression were studied in a multi-site international project in patients with MGUS and SMM with available serial laboratory values in the Precursor Asymptomatic Neoplasms by Group Effort Analysis (PANGEA) model published in 2023 [150]. Patient age, serum FLC ratio, M-protein concentration in g/dL, creatinine concentration in mg/dL, and hemoglobin values over time were used in the model with and without bone marrow plasma cell percentages. Significantly, these models were more accurately predictive for individual patients than the existing models, even in the absence of a bone marrow biopsy using only the other laboratory values, and a freely accessible online user-friendly tool is available for application [150]. Additional cohorts representing the African American population that have a higher prevalence of MGUS (reviewed in [151]), and circulating tumor cells, cell-free DNA, and immune variables, are planned to be studied in this model [150]. Further, at least three randomized clinical trials have shown a survival benefit by starting treatment at the SMM stage [22,152,153]. The reader is referred to the European Myeloma Network consensus statement [154] and the cited review [155] for guidelines and ongoing clinical trials for treating patients with SMM.

Genomics in newly diagnosed multiple myeloma

Multiple studies performed in the last two decades using gene expression profiling, copy number arrays, and next-generation sequencing have shown that MM is genetically and clinically heterogeneous (reviewed in [156,157]). The mitogen-activated protein kinase (MAPK) pathway genes KRAS, NRAS, and BRAF are the most frequently mutated in MM, followed by genes in the nuclear factor κ-light chain-enhancer of activated B cells (NF-κB) pathway [158-160]. However, these mutations had no impact on survival in a large clinical trial in 2015 [158]. In that same clinical trial, mutations in TP53, ATM, ATR, CCND1 and del(17p), and 1q amplification conferred an adverse impact on OS, while ISS-III, age >70 years, t(4;14), MYC translocations, and mutations in TP53, ATM, ATR, and ZFHX4 were prognostic for PFS [158]. In a study of 1,273 NDMM samples in the Myeloma Genome Project with available clinical, outcome, and exome sequencing data, 63 driver genes with mutations were identified, of which 17 (27%) genes were potentially actionable [161], but only the mutated TP53 driver gene was an indicator of clinical outcome [162]. The study noted dependencies between oncogenic mutations and translocation subgroups, suggesting that the path to progression from the initial primary initiating events in myeloma is nonrandom [161,162]. Multiple types of driver alterations were identified: (1) mutated genes, (2) chromosomal translocations, (3) chromosomal and segmental losses and gains, (4) loss of heterozygosity, and (5) an APOBEC (Apolipoprotein B mRNA Editing Catalytic polypeptide-like) deaminase [163] mutational signature, which leads to C>T/G>A transitions in specific trinucleotides. Importantly, an increased number of driver alterations had a poor prognosis [161,162], a concept validated further by the following examples:

In the Myeloma Genome Project, bi-allelic TP53 inactivation or amplification (≥4 copies) of CKS1B (chromosomal location 1q21), comprising 6.1% of the cohort identified a higher-risk subgroup within the ISS-III patients with poor survival [164].

The dual combination of t(4;14) and mutated TP53 showed the shortest OS in another study [165].

A classifier using β2-microglobulin and serum LDH with mRNA levels of APOBEC2, APOBEC3B, and inflammation-related genes IL11, TGFB1, and TGFB3 further stratified NDMM patients in the highest R2-ISS risk group [166].

A recent study showed chromosomal 1q21 gain or amplification combined with mutated KRAS in patients with NDMM to predict a higher risk of developing extramedullary disease [167].

Importantly, in 2019, complex structural abnormalities, including chromothripsis, defined as “the catastrophic shattering of a chromosome followed by re-ligation in a random order” [168], and a novel replication-based mechanism of templated insertions were identified as major initiating driver events [169]. These genetic events and chromosomal hyperdiploidy occur as early drivers of disease [169]. In contrast, point mutations in genes, genomic duplication, and chromoplexy, defined as “intricately weaved genomic rearrangements occurring in concert” [170], occur later in the disease [169].

In the prospective CoMMpass (Relating Clinical Outcomes in Multiple Myeloma to Personal Assessment of Genetic Profile) study comprising 1,143 patients with previously untreated NDMM from the USA, Canada, Spain, and Italy diagnosed between 2011 and 2016, high risk was defined by one or more of the following genetic features: del(17p13), t(14;16), t(14;20), t(8;14), and t(4;14) [171]. This study described five hyperdiploid and three non-hyperdiploid subtypes of NDMM, with no difference in outcomes between the hyperdiploid and non-hyperdiploid subtypes. However, any subtype, regardless of hyperdiploid status and having both 1q gain and loss of chromosome 13 showed inferior overall survival than the other subtypes, with 1q gain attributed as the reason for the inferior outcome [171]. The authors of this study [171] emphasized previous findings [158,161,164] in distinguishing between partial and complete loss-of-function in genes such as TP53 to be clinically important since only a subset of patients with del(17p) detected by conventional cytogenetics analyses are truly high-risk [171].

Further, a 92-gene expression profiling signature that prognosticates NDMM as standard-risk or high-risk for relapse [53], when applied in the Prospective Observational Multiple Myeloma Impact Study (PROMMIS) upgraded or downgraded the risks assigned clinically to impact clinical decisions [172]. The Multiple Myeloma DREAM challenge, which focused on identifying factors predicting high risk in MM, defined as disease progression or death within 18 months of diagnosis, identified a high expression of the PHF19 gene, an epigenetic regulator, as having the strongest association with disease progression [173]. In that study, a prognostic model using age, ISS, and gene expression of only two genes, PHF19 and MMSET (or NSD2) performed well as compared to other complex, gene expression models with higher numbers of genes [173]. In a large single-cell study using multi-omics, chromosome 1q amplification, TP53 mutations, and a high expression of PHF19 were identified as high-risk genetic features associated with relapsed/refractory MM [174].

Genomics in precursor states, MGUS, SMM, and comparison of SMM with NDMM

A whole exome and deep targeted sequencing study of samples from 214 patients with SMM showed that the genetic abnormalities in SMM are already present similar to those in NDMM, except for clinical features that define MM [175]. Structural copy number abnormalities serve as founder events and mutations in gene mutations in cellular pathways lead to disease progression. The following risk factors for progression were identified in this study: KRAS and NRAS single nucleotide variants, deletion 17p, single nucleotide variants of TP53 and ATM genes, and MYC translocations or copy number variations [175].

Interestingly, in another study that compared SMM with MM, KRAS mutations in SMM significantly predicted progression to MM, and mutations in KRAS, NRAS, and FAM46C were considered myeloma-defining events [176]. Significantly fewer mutations in the MAPK and NF-κB pathways were found in cases of SMM, wherein t(11;14)  was the most commonly observed translocation in 23% (n = 19/82) of SMM cases, with 2.4-4.9% of cases harboring the high-risk translocations, t(4;14), t(6;14), and t(14;16) [176]. About 10% (n = 8/82) of SMM cases showed biallelic inactivation of tumor suppressor genes, TP53, DIS3, RB1, FAM46C, and TRAF3, which were significantly fewer than similar events seen in MM [176]. KRAS mutations and del(6q) were associated with a shorter time to progression, but MYC translocations, although identified, were not associated with an adverse prognosis [176]. The number of somatic mutations increased with time in SMM, but the mutational signatures did not change during the progression to multiple myeloma [176].

Further, it was recently shown in both precursor conditions, MGUS and SMM, and cases of MM that there are two types of diseases: a stable disease type that arises later in life and does not progress until decades, and a progressive type that starts earlier in life and progresses early to MM [177]. Significantly, these differences were distinguished despite a low tumor burden characterized by current criteria, but this analysis would require whole genome sequencing [177]. In a smaller genomic analysis of 10 paired samples of SMM and MM, longer and shorter times for the progression of SMM were seen in 6 and 4 cases, respectively [178]. The collective evidence described above suggests that molecular and genomic analyses could identify high risk in MM and identify cases of SMM with a high risk of progression to MM that cannot be identified by current diagnostic and risk stratification criteria.

An individualized risk prediction model integrating clinical and genomic information

In April 2024, based on comprehensive information including clinical, demographic, genomic, and treatment data in 1,933 patients as a training set and 256 patients as a validation set, the first individualized risk prediction model for NDMM (IRMMa) for PFS and OS was described [179]. The model performed better than the existing ISS models, although the numbers of patients were smaller than those in the R-ISS and R2-ISS. The alterations were correlated with gene expression profiling signatures developed at the University of Arkansas [180]. Twenty genomic features, including 1q21 gain or amplification, del(1p), TP53 loss, NSD2 translocations, APOBEC mutational signatures, and copy-number signatures, including chromothripsis, were considered integral for the accuracy of the model. The predictive model is freely available online at the website https://irmma-risk-calculator.miami.edu/. Of note, the model was based on information obtained from a single bone marrow biopsy, and anti-CD38 treatments or MRD as treatment responses were not yet included. As stated by the authors, this model’s accuracy could be improved in the future by integrating information obtained from a peripheral blood sample and evaluating newer treatment options [179].

Figure 6 is created to depict the concepts of progression risks from MGUS and SMM to MM and prognostic risks in MM.

Among all hematological malignancies, the international multiple myeloma community has led to unprecedented advances in the clinical treatments and outcomes of patients with MM along with advancing our understanding of the pathogenesis of the disease and its precursor conditions by working collaboratively toward a common goal of achieving a cure for their patients. These improvements in outcomes have been achieved despite the immense and unique clinical and genetic heterogeneity in this malignant disease arising from plasma cells. The next steps towards a cure include emphasizing the determination of the most precise risks possible by integrating all known intrinsic patient, disease, and extrinsic factors for determining prognostic risk in NDMM and the risk of progression in its precursors, MGUS and SMM, including medullary and extramedullary disease. Reducing obesity and preventing diabetes could prevent and intercept the progression of neoplastic disease from MGUS to SMM or MM. Eliminating barriers to healthcare equity for all patients regardless of ethnic origin, and socioeconomic and geographic factors would help immensely to implement precision medicine.

Also, given the heterogeneity in the biology of MM, including stable and progressive disease biology, and the need to do a bone marrow biopsy to determine MRD and requiring accessibility to high-sensitivity MFC or NGS for MRD evaluation, it would be of value to be able to specify in the future which patients with which specific risk parameters should be evaluated for MRD to be able to make scientifically-based decisions about changing, continuing, or discontinuing therapies based on MRD values. Questions related to MRD assessment that need to be answered include: (1) Does an undetectable MRD value, at whichever level of sensitivity (1 in 10⁴, 10⁵, or 10⁶ cells) available, predict that the disease will not relapse in patients with high-risk myeloma? (2) Similarly, will a detectable or positive MRD value matter in patients with low-risk or standard-risk myeloma? Answers to these questions will help guide the use of MRD when it would be valuable to the management of the patient.

None.

This work received no funding from any source.