Multiple Myeloma and Precursor Plasma Cell Disorders: From Emerging Driver Mutations to Current and Future Therapeutic Strategies

Abstract

Multiple myeloma (MM) is a malignant plasma cell disorder that evolves from precursor conditions including monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM). Understanding the biological continuum and the molecular drivers of disease progression is crucial for early diagnosis and risk-adapted therapy. Recent advances in next-generation sequencing have identified recurrent mutations in the RAS/MAPK, TP53, and MYC pathways, along with epigenetic alterations that contribute to clonal evolution and therapeutic resistance. Novel diagnostic tools including minimal residual disease (MRD) assessment, gene expression profiling, and advanced imaging have improved risk stratification. Therapeutically, the integration of proteasome inhibitors, immunomodulatory drugs, and anti-CD38 monoclonal antibodies has dramatically improved patient outcomes. In parallel, emerging immunotherapies such as CAR-T cells, bispecific T-cell engagers, and antibody–drug conjugates are expanding treatment options, especially in relapsed or refractory settings. Future directions aim to personalize treatment using genomics, target the tumor microenvironment, and leverage synthetic lethality and epigenetic vulnerabilities. This review highlights the evolving landscape of plasma cell disorders from molecular pathogenesis to cutting-edge therapeutic innovations, emphasizing the need for precision medicine approaches to improve survival and quality of life for patients with MM and its precursors.

1. Introduction

Plasma cell dyscrasias exist on a biologic continuum that ranges from premalignant states to overt malignancy. Monoclonal gammopathy of undetermined significance (MGUS) is the most common of these disorders, affecting over 3% of individuals over the age of 50. It is characterized by the presence of a monoclonal protein (M-protein) in the serum at concentrations <3 g/dL and bone marrow plasma cell infiltration <10%, without symptoms or organ damage [1]. Meanwhile, smoldering multiple myeloma (SMM) is an intermediate stage with higher tumor burden than MGUS, defined by ≥3 g/dL serum M-protein and/or 10–60% bone marrow plasma cells, again without myeloma-defining events [2]. Ultimately, multiple myeloma (MM), the malignant endpoint of this spectrum, is defined by ≥10% clonal bone marrow plasma cells or biopsy-proven plasmacytoma with evidence of end-organ damage (hypercalcemia, renal failure, anemia, or bone lesions) or specific biomarkers of malignancy (e.g., ≥60% bone marrow plasma cells, free light chain ratio ≥100, or >1 focal lesion on MRI) [3]. The risk of progression to symptomatic MM is ~1% per year for MGUS and ~10% per year for SMM [4]. MM accounts for approximately 10% of hematologic malignancies and has a global incidence of ~160,000 cases per year. It remains largely incurable despite therapeutic advancements, with significant morbidity from bone disease, renal failure, and infections [5]. The disease disproportionately affects the elderly and African Americans, and survival varies widely based on molecular risk features and response to therapy. Since MM is almost always preceded by MGUS or SMM, early identification and risk stratification are essential for timely intervention. Advances in molecular diagnostics—such as free light chain assays, gene expression profiling, and imaging—have improved our ability to predict progression [6]. Risk-adapted surveillance and early therapeutic strategies in high-risk SMM have shown promise in delaying or preventing progression to MM [7]. This review aims to provide a comprehensive overview of the molecular and clinical evolution from MGUS and SMM to MM. We examine the role of emerging genomic and epigenetic drivers, risk stratification models, and current therapeutic strategies. Furthermore, we explore the landscape of evolving therapies, including immunotherapies and personalized medicine, with the goal of improving outcomes across the spectrum of plasma cell disorders.

2. Pathophysiology and Disease Continuum

2.1. Clonal Evolution from MGUS to SMM to MM

The progression of MM is widely recognized as a multistep clonal evolutionary process beginning with MGUS, advancing to SMM, and eventually transforming into symptomatic MM (Figure 1). This progression is driven by the accumulation of genetic abnormalities, selective pressures, and interaction with the bone marrow microenvironment. Early genomic events such as hyperdiploidy and immunoglobulin heavy chain (IgH) translocations are frequently observed in MGUS and are thought to initiate clonal plasma cell proliferation. As the disease progresses to SMM and MM, secondary alterations—such as mutations in the RAS/MAPK pathway (KRAS, NRAS, BRAF), MYC dysregulation, TP53 deletions, and chromosomal gains (1q) or deletions (13q, 17p)—confer growth and survival advantages to specific subclones [8]. Importantly, clonal heterogeneity is a hallmark of this process. Subclones with varying genomic profiles coexist, and selective pressures—such as microenvironmental interactions or therapeutic intervention—can promote the dominance of specific aggressive subclones [9]. Two main models of evolution have been proposed (

Table 1. Comparison of linear vs. branching clonal evolution in multiple myeloma.

Feature	Linear Evolution	Branching Evolution
Definition	Stepwise accumulation of mutations in a single dominant clone	Multiple subclones diverge from a common ancestor and evolve in parallel
Clonal dynamics	One clone sequentially replaces the previous one (clonal sweeps)	Several clones coexist and compete; dominance can shift over time
Genetic trajectory	Unidirectional with sequential driver mutations	Diverse mutational paths; subclones may acquire different mutations independently
Temporal pattern	Progressive and ordered	Non-linear and dynamic
Driver mutation Distribution	Accumulated in a linear fashion within a single lineage	Found in distinct subclones; can appear independently
Influence of therapy	Selective pressure can accelerate emergence of the next dominant clone	Selective pressure alters subclone proportions; resistant clones may expand
Clinical presentation	Often associated with more aggressive disease progression and shorter survival	More heterogeneous presentation; some subclones may respond differently to therapy
Frequency in MM	Less common (~15–25%)	More common (~60–70%)
Detection	Easier to detect using bulk sequencing due to dominant clone	Requires single-cell sequencing or longitudinal sampling to fully resolve
Example markers	TP53 inactivation, high mutational load	Divergent RAS mutations in separate subclones
Prognostic implication	Often associated with poor prognosis and rapid relapse	Variable prognosis; can enable early detection of resistance pathways
Therapeutic implications	May benefit from aggressive upfront treatment strategies	Supports adaptive or combination therapy approaches to target multiple clones

): linear evolution, where a dominant clone accumulates sequential driver mutations, and branching evolution, where multiple subclones evolve in parallel with distinct mutations and phenotypes [10].

In the linear evolution model (Table 1), a single dominant clone sequentially acquires driver mutations over time, resulting in a unidirectional path toward increasingly aggressive disease. Each new subclone replaces the previous one, creating a dominant “clone of the moment” with a continuous lineage. This model implies a stepwise pattern of progression, often associated with highly aggressive disease and poor prognosis, especially when driven by mutations such as del(17p) or TP53 inactivation [11]. Linear evolution tends to be less influenced by the microenvironment and more by internal genomic instability, leading to rapid clonal sweeps. Patients with linear evolution tend to experience shorter survival post-relapse compared to other evolutionary patterns. In a longitudinal analysis, the median overall survival (OS) for patients with linear evolution was only 25 months, significantly worse than for those with branching or stable patterns [12]. In contrast, the branching evolution model describes a Darwinian diversification process in which multiple subclones evolve in parallel from a common ancestral clone, each acquiring distinct sets of mutations (Table 1). These subclones coexist and compete, often shaped by microenvironmental factors, therapy-induced selective pressures, and immune evasion mechanisms [9]. Branching evolution is the predominant mode observed in MM. In a whole-exome sequencing (WES) study of paired diagnostic and relapse samples, over 70% of patients showed evidence of branching evolution, often with multiple founding clones and fluctuating subclonal dynamics [13]. Importantly, some subclones that are minor at diagnosis may later dominate at relapse, indicating selection under treatment pressure. Branching evolution allows for parallel development of resistance, as different subclones may harbor distinct sensitivities. For instance, one clone may respond to proteasome inhibitors (PIs), while another may resist and expand during therapy, eventually driving relapse [14]. Branching evolution also underlies the emergence of extramedullary disease and minimal residual disease (MRD) escape variants. As such, linear evolution may justify early aggressive treatment to prevent the accumulation of deleterious mutations, whereas branching evolution supports adaptive or combination strategies targeting multiple clones and pathways. Furthermore, branching evolution highlights the need for dynamic risk stratification, as static baseline markers may miss evolving resistance pathways. Clinical trials increasingly aim to incorporate serial molecular assessments to guide therapy selection based on evolving clonal architecture [15].

Furthermore, WES and whole-genome sequencing (WGS) of paired samples from MGUS/SMM and later MM stages demonstrate that many of the driver mutations seen in MM are already present at the precursor stages. For example, NRAS, BRAF, and DIS3 mutations and high-risk cytogenetic abnormalities like t(4;14) and del(17p) may be detected in MGUS or SMM albeit in smaller clonal populations [16]. This suggests that clonal stability rather than rapid mutation accumulation may define many cases of progression, with certain dominant subclones simply expanding over time [17]. Nevertheless, despite the presence of early genomic lesions, disease progression is not solely driven by intrinsic mutations. The bone marrow microenvironment—comprising stromal cells, cytokines, immune components, and extracellular matrix—plays a crucial role in supporting malignant plasma cell survival, immune evasion, and therapy resistance [18]. Additionally, immune dysregulation and loss of immune surveillance, particularly T-cell exhaustion and altered cytokine profiles, contribute to clonal dominance and disease advancement [19]. Thus, the evolution from MGUS to MM is not a sudden transformation but rather a gradual process marked by genetic diversification, subclonal competition, and microenvironmental adaptation.

2.2. Microenvironmental Influences

As mentioned above, the progression from MGUS and SMM to overt MM is not solely dictated by tumor-intrinsic genetic alterations but is profoundly influenced by the bone marrow microenvironment [2]. The bone marrow niche, composed of stromal cells, osteoblasts, osteoclasts, endothelial cells, mesenchymal stem cells (MSCs), immune cells, and the extracellular matrix, provides essential cues that support plasma cell survival, proliferation, and immune evasion (

Table 2. Components of the bone marrow microenvironment involved in multiple myeloma.

Component	Cell Type/Molecule	Role in MM Pathogenesis	Key Molecular Mechanisms/Pathways
Stromal cells	Bone marrow stromal cells (BMSCs)	Promote MM cell survival, proliferation, and drug resistance	IL-6, SDF-1/CXCL12 → CXCR4, adhesion molecules (VCAM-1, VLA-4), NF-κB activation
Osteoclasts	Multinucleated bone-resorbing cells	Promote bone destruction and release growth factors for MM cells	RANK/RANKL, M-CSF, IL-6, TGF-β; release of IGF-1 and calcium enhances MM growth
Osteoblasts	Bone-forming cells	Suppressed in MM, leading to bone lesions	Inhibition via DKK1, sFRP2, and activin A; Wnt pathway inhibition
Endothelial cells	Vascular endothelial cells	Promote angiogenesis and MM progression	VEGF, FGF-2, HIF-1α pathways; MM cells stimulate angiogenic switch
Immune cells—Myeloid lineage	Myeloid-derived suppressor cells (MDSCs)	Suppress anti-MM immunity, support MM cell survival	Arginase-1, ROS, IL-10, TGF-β; inhibit cytotoxic T cells and NK cells
Immune cells—Lymphoid lineage	Regulatory T cells (Tregs)	Inhibit effective immune responses against MM	IL-10, TGF-β, CTLA-4; suppression of CD8+ T-cell activation
Dendritic cells	Antigen-presenting cells	Impaired antigen presentation; promote immune tolerance in MM	Downregulation of costimulatory molecules, IL-6, IDO expression
Macrophages	Tumor-associated macrophages (TAMs)	Support MM growth and immune evasion	IL-6, TNF-α, CD163 expression; M2 polarization promotes tumor-supportive environment
Mesenchymal stem cells (MSCs)	Multipotent stromal progenitors	Give rise to BMSCs; modulate inflammation and MM niche	Secretion of IL-6, CXCL12; direct interactions with MM cells via adhesion molecules
Adipocytes	Bone marrow fat cells	Provide metabolic support and cytokines for MM cells	Leptin, adiponectin, IL-6, lipolysis-derived fatty acids; affect metabolism and resistance
Extracellular matrix (ECM)	Fibronectin, collagen, laminin, hyaluronan	Promotes MM adhesion, survival, and drug resistance	Integrins (VLA-4/VCAM-1), FAK, PI3K/AKT, ERK signaling; CAM-DR (cell adhesion-mediated drug resistance)
Cytokines and chemokines	IL-6, IL-10, TGF-β, SDF-1 (CXCL12), TNF-α	Mediate pro-survival, pro-migratory, and immunosuppressive signals	Activation of JAK/STAT3, MAPK, NF-κB, and PI3K/AKT pathways
Growth factors	VEGF, IGF-1, BAFF, APRIL, FGF	Enhance angiogenesis, MM cell proliferation, and immune evasion	Binding to receptors on MM cells (e.g., IGF-1R, VEGFR), MAPK/ERK and AKT activation
Exosomes/extracellular vesicles	Secreted by MM cells and stromal cells	Mediate intercellular communication, resistance, and immune modulation	Transfer of miRNAs, proteins; enhance angiogenesis, suppress T-cell function

) [20]. A central mechanism of progression is the remodeling of the stromal and endothelial compartments, which begins during MGUS. MSCs show early transcriptional changes that promote inflammation, while endothelial cells shift toward an angiogenic phenotype in MM, enhancing tumor cell access to nutrients and promoting dissemination [21]. Additionally, disruption of the endosteal niche—marked by increased osteoclast activity and loss of osteoblast function—facilitates tumor proliferation and reduces immune surveillance [22]. The immune microenvironment plays a critical role in allowing malignant plasma cells to escape host defenses. MM cells secrete immunosuppressive cytokines (e.g., TGF-β, IL-10) and interact with regulatory T cells, myeloid-derived suppressor cells, and plasmacytoid dendritic cells to create a tolerogenic and tumor-promoting milieu [23]. Natural killer (NK) cell function becomes increasingly impaired during disease progression, particularly via expansion of dysfunctional CD56^bright NK cell subsets with low cytotoxicity [24]. Furthermore, hypoxic conditions in the BM niche induce adaptive responses in MM cells, including metabolic reprogramming, resistance to apoptosis, and enhanced immune evasion through altered antigen presentation and immune checkpoint expression [25]. These adaptations are essential for the persistence of MRD and disease relapse, even after deep therapeutic responses. Collectively, these findings highlight that MM is not a disease solely of rogue plasma cells but also of a corrupted microenvironment that enables tumor survival and immune escape.

2.3. Hallmarks of Malignant Plasma Cell Transformation

The transformation of normal plasma cells into malignant counterparts seen in MM involves a complex interplay of genomic, epigenetic, and microenvironmental alterations (Figure 2). This process aligns with many of the canonical “hallmarks of cancer” and manifests uniquely in the plasma cell lineage. One fundamental hallmark is genomic instability, which drives clonal evolution and progression from MGUS to MM. Malignant plasma cells accumulate chromosomal aberrations—such as hyperdiploidy, IgH translocations (e.g., t(11;14), t(4;14)), and deletions of 13q or 17p—that increase over time and correlate with disease severity. These genetic insults promote cell survival, therapy resistance, and immune evasion [26]. Another critical feature is dysregulated proliferation and evasion of apoptosis, sustained by signaling through IL-6, NF-κB, and PI3K/AKT pathways. Malignant cells upregulate anti-apoptotic proteins such as BCL-2 and MCL-1, fostering survival even under genotoxic stress [27]. Immune evasion is also a hallmark. MM cells reduce antigen presentation, downregulate costimulatory molecules, and recruit immunosuppressive cell populations like regulatory T cells and myeloid-derived suppressor cells. Additionally, hypoxic adaptation in the bone marrow niche induces metabolic and transcriptional shifts that promote resistance to cytotoxic immune cells [2,25]. Epigenetic reprogramming contributes to plasma cell transformation by modifying gene expression independently of DNA sequence. Long non-coding RNAs (lncRNAs) regulate key pathways involved in proliferation, apoptosis, and cell cycle control in MM. Dysregulated expression of lncRNAs, such as MALAT1 and NEAT1, has been implicated in disease progression and drug resistance [28].

MM cells also exhibit a distinct metabolic phenotype characterized by glutamine addiction, a feature that has garnered increasing attention as a potential therapeutic vulnerability. Multiple studies collectively demonstrate that MM cells exhibit a profound dependence on extracellular glutamine (Gln)—a metabolic vulnerability with promising therapeutic implications. Bolzoni et al. showed that MM cells and patient-derived CD138(+) plasma cells produce high levels of ammonium and glutamate in Gln-rich conditions and exhibit high GLS1 (glutaminase) and low GS (glutamine synthetase) expression, confirming their reliance on extracellular Gln. They also identified ASCT2 as the key Gln transporter, where its inhibition significantly reduced MM cell viability in vitro and in vivo [29]. Bajpai et al. further highlighted how Gln withdrawal induces BIM expression and sensitizes MM cells to venetoclax, especially by shifting BIM binding toward BCL-2, suggesting a synthetic lethal strategy involving glutaminase inhibition and BCL-2 blockade [30]. Supporting this, Soncini et al. demonstrated that L-asparaginase (ASNase), a glutaminolytic enzyme, synergizes with carfilzomib by exacerbating mitochondrial stress, ROS generation, and DNA repair disruption, sensitizing even carfilzomib-resistant MM cells [31]. Finally, Effenberger et al. confirmed that glutamine deprivation or GLS1 inhibition with compound 968 induced rapid MYC degradation and apoptosis, especially in MYC-expressing MM cells, while glucose deprivation did not elicit the same apoptotic response [32]. Together, these findings reveal that targeting glutamine metabolism—via transporter blockade, enzymatic depletion, or glutaminase inhibition—may be a potent and selective therapeutic strategy, particularly in combination with venetoclax or PIs.

In addition to metabolic reprogramming, aberrant cellular adhesion and migration are features that enable malignant plasma cells to colonize the bone marrow and form lytic lesions. MM cells display functional diversity in their adhesion properties, with subpopulations exhibiting distinct integrin profiles, migratory behavior, and interaction with extracellular matrix components [33]. Collectively, these hallmarks—genomic instability, apoptotic resistance, immune evasion, epigenetic dysregulation, metabolic reprogramming, and altered adhesion—enable the malignant transformation and expansion of plasma cells, ultimately driving the pathogenesis of multiple myeloma.

3. Genomic and Epigenetic Landscape

3.1. Common Cytogenetic Abnormalities

3.1.1. Hyperdiploidy

Hyperdiploidy, defined as the presence of more than the normal diploid number of chromosomes (typically 47–75), is one of the most common cytogenetic abnormalities in MM, occurring in approximately 45–60% of newly diagnosed patients. It is characterized by trisomies of specific odd-numbered chromosomes, most commonly 3, 5, 7, 9, 11, 15, 19, and 21 [34]. Hyperdiploidy is typically considered an early and initiating event in myelomagenesis, often seen even at the MGUS stage [34]. This cytogenetic pattern defines a distinct molecular subgroup of MM known as “hyperdiploid MM” (H-MM), which is associated with lower rates of IgH translocations and generally more favorable clinical outcomes [35,36]. Several studies have shown that hyperdiploidy confers a survival advantage, particularly in patients lacking high-risk abnormalities such as t(4;14), t(14;16), del(17p), or gain of 1q. For example, in a cohort analysis, hyperdiploid patients had improved PFS compared to non-hyperdiploid counterparts (hazard ratio [HR] 1.9) [37]. However, when hyperdiploidy coexists with high-risk cytogenetic lesions, its protective effect may be mitigated or lost entirely [38]. At the molecular level, H-MM is marked by the upregulation of genes related to protein biosynthesis, ribosomal function, and oxidative phosphorylation, particularly on the gained chromosomes. In a seminal study by Agnelli et al., gene expression profiling of 66 newly diagnosed MM cases revealed that H-MM displayed significant upregulation of 204 genes, the majority of which mapped directly to the gained chromosomes. These genes were functionally linked to translation, transcriptional machinery, and mitochondrial energy production, indicating a gene dosage effect driven by the extra chromosomal content [39]. Further supporting this, a follow-up analysis identified that ribosomal proteins and oxidative phosphorylation components were consistently overexpressed in H-MM, suggesting that this subtype maintains a metabolically active and protein-synthesis-intensive phenotype, potentially reflecting a less aggressive, more differentiated state [40]. These molecular features differentiate it from non-H-MM (NH-MM), which more often involves IgH translocations and cell cycle dysregulation (

Table 3. Comparison of hyperdiploid vs. non-hyperdiploid multiple myeloma.

Feature	Hyperdiploid Multiple Myeloma (H-MM)	Non-Hyperdiploid Multiple Myeloma (NH-MM)
Definition	Presence of trisomies (usually odd-numbered chromosomes)	Characterized by structural abnormalities, especially IgH translocations
Common cytogenetic features	Trisomies of chromosomes 3, 5, 7, 9, 11, 15, 19, 21	IgH translocations: t(11;14), t(4;14), t(14;16), t(14;20), t(6;14)
Frequency	~50–60% of newly diagnosed MM	~30–40% of newly diagnosed MM
Oncogenic mechanism	Gene dosage effect from trisomies	Oncogene activation via enhancer hijacking (e.g., CCND1, FGFR3/MMSET, MAF)
Gene expression profile	Enrichment in genes linked to protein synthesis and oxidative phosphorylation	Enrichment in cell cycle, proliferation, and oncogenic transcription factors
MYC deregulation	Less frequent	More frequent, especially in t(4;14), t(14;16) subtypes
Cyclin D deregulation	Via gene dosage (CCND1, CCND2)	Via IgH translocations (e.g., t(11;14) → CCND1 overexpression)
Prognosis	Generally more favorable	Variable, often poorer (especially with t(4;14), del(17p), or 1q gain)
Associated risk abnormalities	Fewer high-risk features	More frequent del(17p), 1q gain, and complex karyotypes
Response to therapy	Good response to standard therapy	Often requires risk-adapted or novel agents (e.g., bortezomib in t(4;14))
Microenvironmental interaction	Less dependent on niche signaling	Greater dependence on IL-6 and stromal support in early clones
Progression pattern	Typically slower progression from MGUS	Can have rapid evolution from SMM or de novo aggressive disease
Common clinical features	Older age, indolent course	Younger patients more common; higher tumor burden and extramedullary disease

). NH-MM shows enrichment for genes involved in cell cycle regulation, proliferation, and oncogene activation (e.g., FGFR3, MMSET, MAF, and MYC), consistent with its more aggressive behavior and poorer prognosis [41]. Moreover, NH-MM exhibits a higher frequency of chromosomal deletions, such as del(13q), del(17p), and del(1p), which are linked to resistance and inferior outcomes. Interestingly, despite their differing genomic origins, both H-MM and NH-MM may converge on shared oncogenic pathways. For example, the deregulation of cyclin D genes—via gene dosage in H-MM and IgH translocations in NH-MM—suggests a unified mechanism of cell cycle dysregulation in MM pathogenesis [42]. Additionally, emerging data suggests that the timing of hyperdiploidy is early, often arising decades before clinical diagnosis. A WGS study revealed that hyperdiploid changes may begin in early adulthood and require years to complete, accumulating secondary mutations during disease latency [43].

3.1.2. IgH Translocations

Translocations involving the IgH locus at chromosome 14q32 are among the most frequent and clinically significant chromosomal alterations in MM. These rearrangements juxtapose oncogenes to the highly active IgH enhancer, leading to deregulated gene expression that drives myelomagenesis. The most common recurrent IgH translocations are t(11;14)(q13;q32), t(4;14)(p16;q32), and t(14;16)(q32;q23), each with distinct biological and clinical implications (

Table 4. IgH translocations in multiple myeloma.

Translocation	Partner Gene/Locus	Frequency	Molecular Consequence	Associated Clinical Features/Risk
t(11;14)(q13;q32)	CCND1 (cyclin D1)	~15–20%	Overexpression of cyclin D1, promoting G1/S cell cycle entry	Standard-risk; enriched in light chain amyloidosis, lymphoplasmacytic morphology, and BCL-2 dependency (venetoclax-responsive)
t(4;14)(p16;q32)	FGFR3/MMSET (NSD2)	~15%	Dual activation: FGFR3 (MAPK signaling) and MMSET (epigenetic modulation)	High-risk; associated with poor prognosis, early relapse, and benefit from bortezomib-based therapy
t(14;16)(q32;q23)	MAF (v-maf musculoaponeurotic fibrosarcoma)	~5–7%	Upregulation of MAF transcription factor, altering adhesion and migration genes	High-risk; associated with extramedullary disease and plasma cell leukemia-like features
t(14;20)(q32;q12)	MAFB	<2%	Overexpression of MAFB, disrupting plasma cell differentiation	High-risk; rare but aggressive, with poor response to conventional therapy
t(6;14)(p21;q32)	CCND3 (cyclin D3)	<1–2%	Overexpression of cyclin D3, promoting cell cycle progression	Rare; clinical significance not fully established; intermediate prognosis
t(8;14)(q24;q32)	MYC (secondary event)	~15–30% (progression)	Enhancer hijacking leads to MYC overexpression	Secondary translocation in disease progression; associated with high tumor burden and relapse
t(9;14)(p13;q32)	PAX5	Very rare	Disrupts B-cell differentiation pathway	Extremely rare in MM; more common in lymphomas

) [2]. The t(11;14) translocation, present in 15–20% of MM patients, involves the cyclin D1 gene (CCND1), promoting cell cycle progression. It is frequently associated with a lymphoplasmacytic morphology and light chain-only subtype. While traditionally considered a standard-risk abnormality, recent evidence reveals clinical heterogeneity depending on coexpressed markers like CD20 and BCL-2 dependency. Importantly, t(11;14)-positive MM exhibits heightened sensitivity to the BCL-2 inhibitor venetoclax, representing the first predictive biomarker in MM therapy [44,45]. The t(4;14) translocation, found in ~15% of MM cases, results in overexpression of FGFR3 and MMSET (NSD2) and is considered high-risk. It is associated with poor response to standard therapies and shortened progression-free survival (PFS) and OS, even in the era of PIs [46,47]. The poor prognosis is thought to be mediated by enhanced oncogenic signaling, epigenetic dysregulation, and chromatin remodeling initiated by MMSET. Next, the t(14;16) translocation is rarer (~4–5%) and leads to overexpression of the MAF oncogene. This lesion is also categorized as high-risk and frequently coexists with additional adverse cytogenetic changes like del(17p) and gain(1q), further worsening prognosis. The pathogenic effects of MAF overexpression include increased cell adhesion, migration, angiogenesis, and deregulated transcriptional programs. Furthermore, t(14;16) MM often harbors an APOBEC mutational signature, contributing to genomic instability and aggressive disease behavior [48,49]. Interestingly, patients with IgH translocations involving undefined partners (t(14;unknown)) have been shown to exhibit favorable survival outcomes in some cohorts, particularly in the context of bortezomib-based regimens, highlighting the diversity and prognostic complexity of IgH rearrangements [50].

3.1.3. Chromosome 13q and 17p Deletions

Deletions of chromosome 13q [del(13q)] are among the most frequently observed chromosomal abnormalities in MM, affecting approximately 50–60% of patients at diagnosis when using sensitive fluorescence in situ hybridization (FISH) techniques [51]. These deletions often involve large segments of the long arm of chromosome 13, typically including the RB1 gene at 13q14, and are frequently associated with monosomy 13. While once considered a uniformly high-risk feature, the prognostic value of del(13q) depends on detection method and co-occurrence with other abnormalities. For instance, deletions detected by metaphase cytogenetics confer a poorer prognosis compared to those detected solely by interphase FISH [52]. Del(13q) is commonly found in combination with other high-risk cytogenetic events, such as t(4;14) and del(17p), and may reflect underlying genomic instability. Moreover, it is more frequently associated with hypodiploid karyotypes and a poor response to standard therapy, especially in patients not treated with PIs [53]. However, novel agents like bortezomib appear to partially overcome the adverse impact of del(13q) [54].

Deletion of chromosome 17p [del(17p)], involving the TP53 tumor suppressor gene at 17p13, is a well-established high-risk marker in MM, present in about 10% of newly diagnosed cases. The prognostic effect of del(17p) is especially pronounced when more than 60–70% of plasma cells harbor the deletion. It is independently associated with shorter PFS and OS, even with high-dose therapy and novel agents [55]. In fact, del(17p) is frequently coexistent with TP53 mutations, and the combined inactivation of both alleles contributes to an aggressive clinical course, marked by drug resistance and rapid relapse [56]. Importantly, the adverse impact of del(17p) cannot be fully mitigated by current treatments, highlighting the urgent need for novel therapeutic approaches, including TP53-independent strategies and enhanced immune-based interventions. Current guidelines recommend integrating del(17p) status into all major risk stratification systems and clinical trial designs.

3.2. Emerging Driver Mutations

3.2.1. RAS Pathway Mutations

Activating mutations in the RAS-RAF-MAPK pathway—specifically in KRAS, NRAS, and BRAF—are among the most frequently observed somatic alterations in MM (

Table 5. Comparison of RAS pathway mutations in multiple myeloma.

Feature	KRAS Mutation	NRAS Mutation	BRAF Mutation
Gene location	Chromosome 12p12.1	Chromosome 1p13.2	Chromosome 7q34
Frequency in MM	~20–25%	~20–25%	~4–8%
Common mutation hotspots	G12D, G12V, G13D, Q61H	Q61K, Q61R, G12D, G13R	V600E (most common), K601N
Activation effect	Constitutive activation of MAPK and PI3K/AKT	Preferential activation of MAPK signaling	Strong MAPK pathway activation via MEK/ERK
Pathway preference	Both MAPK and PI3K pathways	MAPK (ERK-driven proliferation)	MAPK pathway exclusively
Mutual exclusivity	Generally mutually exclusive with NRAS/BRAF	Rare co-occurrence with KRAS/BRAF	Usually exclusive of RAS mutations
Prognostic impact	Variable; may be associated with relapse	May be associated with early progression	V600E associated with poor prognosis and high-risk disease
Clonal distribution	Often subclonal at diagnosis, may become clonal at relapse	Often clonal and persistent through disease course	Can be early or late event
Functional role in MM	Promotes cell proliferation, resistance to apoptosis	Drives proliferation and survival	Promotes rapid proliferation and metabolic changes
Response to MEK inhibitors	Variable; better response in dual PI3K/MEK inhibition	Similar to KRAS; requires combination therapy	More responsive to BRAF/MEK inhibitors (e.g., vemurafenib + cobimetinib)
Preclinical targetability	Requires combined targeting of PI3K and MEK	Limited success with MEK inhibitors alone	V600E mutation shows sensitivity to BRAF inhibitors in vitro
Clinical trial implications	Enriched in biomarker-driven MEK/AKT inhibitor studies	Evaluated in MAPK pathway-targeted trials	Enrolled in BRAF inhibitor trials [64] (e.g., vemurafenib [VEM] + cobimetinib [COBI])

), found in approximately 40–55% of newly diagnosed cases [57]. These mutations contribute to aberrant downstream signaling that promotes tumor cell survival, proliferation, and resistance to therapy. KRAS and NRAS mutations are the most common, affecting codons G12, G13, and Q61. These mutations tend to be mutually exclusive and often occur in subclonal populations, highlighting their role in intratumoral heterogeneity and disease evolution [58]. Notably, NRAS mutations—particularly at codon Q61—have been associated with resistance to PIs like bortezomib, leading to inferior clinical responses and shorter time to progression [59]. BRAF mutations, though less frequent (observed in 4–15% of cases), primarily involve the V600E variant. This mutation activates the mitogen-activated protein kinase (MAPK) pathway independently of upstream signals and has been associated with more aggressive disease and shorter PFS in some cohorts [60]. BRAF-mutant MM cells show increased proteasome activity and reduced ER stress, contributing to drug resistance and survival advantage [61]. Functionally, these mutations converge on the MAPK signaling cascade, promoting phosphorylation of downstream effectors such as MEK and ERK. This cascade enhances proteasome capacity, reduces unfolded protein response activation, and distances malignant cells from apoptosis [61]. Importantly, pharmacologic inhibition of BRAF (e.g., with vemurafenib) or MEK can resensitize MM cells to PIs, especially in those harboring BRAF V600E or activating RAS mutations [62]. Clinically, patients with RAS/MAPK pathway mutations tend to have a higher tumor mutational burden and more rapid progression, particularly in the relapsed/refractory setting [63]. Thus, mutation testing for KRAS, NRAS, and BRAF is becoming increasingly relevant for prognostication and for guiding the use of combination targeted therapies, such as PIs with MEK or BRAF inhibitors.

3.2.2. TP53 and DNA Repair Genes

Mutations and deletions involving the TP53 gene, located on chromosome 17p13, are among the most critical high-risk genetic events in MM. TP53 functions as a tumor suppressor, regulating key cellular processes such as DNA repair, apoptosis, and cell cycle arrest. Deletion of 17p (del(17p)), which encompasses TP53, occurs in approximately 10% of newly diagnosed MM patients, and TP53 mutations often co-occur in this subset, compounding the adverse prognosis [56]. These mutations are most frequently located in the DNA-binding domain and are primarily missense changes that disrupt TP53’s transcriptional activity. Patients with TP53 mutations in the context of del(17p) exhibit significantly shorter PFS and OS, with studies reporting median survival as low as 4 months in high-clonality cases [65]. Gene expression profiling has revealed that low TP53 expression correlates with a unique signature of deregulated genes involved in apoptosis, DNA repair, and cell cycle regulation, forming a distinct high-risk subgroup [66]. In parallel, mutations in DNA damage response (DDR) and DNA repair genes are increasingly recognized as key contributors to MM pathogenesis and resistance to therapy. Alterations in genes involved in homologous recombination (e.g., ATM, BRCA1/2), non-homologous end joining, and base excision repair lead to genomic instability and clonal evolution [67]. Notably, circulating tumor DNA (ctDNA) analyses have shown that DNA repair gene mutations are more prevalent and clonally expanded in relapsed/refractory MM (RRMM) than at diagnosis, with high mutational burden predicting poorer OS [68]. Targeting these DNA repair vulnerabilities is a growing therapeutic focus. For example, CDK12 inhibition with THZ531 suppresses DNA repair gene expression and induces synthetic lethality when combined with PARP (e.g., olaparib) or DNA-PK inhibitors (KU-0060648), offering promising preclinical results in MM models [69].

3.2.3. MYC Deregulation

Deregulation of the MYC oncogene, located on chromosome 8q24, is a pivotal event in the pathogenesis and progression of MM. Unlike primary translocations seen at diagnosis (e.g., IgH translocations), MYC abnormalities typically arise during the transition from SMM to symptomatic disease, making them one of the earliest and most consistent secondary genetic events in MM evolution [70]. Approximately 30–50% of MM cases harbor structural variants involving MYC, including translocations with immunoglobulin enhancers (e.g., IgH, IgK, IgL), amplifications, and complex rearrangements. These events lead to aberrant MYC overexpression and transcriptional activation of downstream targets involved in metabolism, proliferation, and survival [71]. Importantly, the presence of MYC rearrangements correlates with increased tumor burden and poor clinical outcomes, particularly in relapsed or high-risk MM [72]. Beyond structural alterations, recent studies have revealed epigenetic mechanisms driving MYC activation. In cases lacking MYC rearrangements, enhancer elements with increased chromatin accessibility—bound by MM-specific transcription factors such as IRF4 and MAF—can elevate MYC expression via a super-enhancer mechanism. These regulatory elements are enriched in active histone marks (e.g., H3K27ac) and are found to be amplified in a subset of MM patients, contributing to MYC overexpression without genomic rearrangement [73]. Functionally, MYC acts as a transcriptional amplifier, upregulating genes involved in ribosome biogenesis, glucose metabolism, cell cycle progression, and suppression of apoptosis. MYC also cooperates with chromatin regulators and lncRNAs like PVT1, located adjacent to the MYC locus, further stabilizing its oncogenic activity [74]. Importantly, MYC deregulation is tightly associated with disease progression. Its overexpression distinguishes MM from MGUS and SMM and contributes to drug resistance, including reduced sensitivity to immunomodulatory drugs and PIs [75]. Inhibition of MYC activity through direct (e.g., c-Myc/Max disruptors) or indirect approaches (e.g., BET inhibitors like JQ1) has shown efficacy in preclinical MM models by reversing MYC-driven transcriptional programs and inducing tumor regression [76].

3.2.4. Mutational Signatures and Timing in Progression

The transformation from MGUS to MM involves the sequential accumulation of genetic alterations shaped by distinct mutational processes. These processes leave behind characteristic patterns—mutational signatures—that can be used to reconstruct the temporal and mechanistic landscape of disease evolution. Recent WGS studies have identified key mutational signatures that dominate at various phases of MM development. A landmark study analyzing 89 WGS and 973 exomes revealed that many MM-related mutational events are acquired decades before diagnosis, with some key chromosomal gains occurring in early adulthood, suggesting a prolonged subclinical evolution [77]. This early phase is dominated by age-related signatures (SBS1, SBS5) and activation-induced cytidine deaminase (AID) activity (SBS9), indicating their foundational role in initiating clonal plasma cell expansion. As disease progresses toward symptomatic MM, additional mutational processes emerge. In particular, APOBEC-mediated mutagenesis (SBS2 and SBS13) becomes prominent and is associated with disease progression, subclonal diversification, and poor clinical outcomes [78]. APOBEC mutations are especially enriched in RRMM and contribute to genomic instability. This pattern supports a two-phase model: early transformation is driven by AID and age-related mutations, while late progression is fueled by APOBEC activity and other stress-related mutational processes. A broader mutational burden—termed tumor mutational burden (TMB)—also increases significantly from MGUS to MM, particularly in non-synonymous variants and 3′/5′ UTR regions, which may disrupt regulatory mechanisms. High TMB correlates with APOBEC activity and is linked to poorer survival outcomes [79]. Notably, these mutational processes operate in a clonally structured manner. Clonal mutations (early events) are enriched for AID and aging signatures, whereas subclonal mutations (later events) reflect APOBEC and additional novel signatures, such as MM-1 and Signature #8—indicating evolving mutational mechanisms during progression [78]. Finally, longitudinal studies of paired MGUS/SMM and MM samples confirm that most driver mutations and chromosomal changes are already present at the precursor stage. However, their relative clonal abundance changes over time, suggesting that clonal expansion rather than new driver acquisition is often responsible for the transition to symptomatic disease [80].

3.3. Epigenetic Dysregulation

Epigenetic dysregulation is a defining feature of MM pathogenesis and progression [2]. Among the major epigenetic mechanisms, histone modifications and DNA methylation play pivotal roles in regulating chromatin accessibility and gene expression. These changes contribute to both early plasma cell transformation and the development of drug resistance and clonal evolution during disease progression. Histone modifications, such as methylation and acetylation, are catalyzed by histone-modifying enzymes and dynamically control gene transcription through chromatin remodeling. In MM, enzymes such as MMSET (WHSC1), which is overexpressed due to the t(4;14) translocation, are critical in driving tumorigenesis via increased H3K36 methylation, altering chromatin states and transcriptional programs in favor of malignant proliferation [81]. Additionally, UTX, a histone demethylase responsible for removing repressive marks (e.g., H3K27me3), is frequently inactivated in MM via mutations, deletions, or promoter hypermethylation, leading to abnormal gene silencing and worse outcomes [82]. Beyond individual enzymes, global patterns of histone methylation can be detected even in circulating nucleosomes, and levels of H3K9me1—a mark associated with heterochromatin and gene repression—are significantly altered in MM patients’ plasma, suggesting systemic changes in chromatin regulation [83]. DNA methylation, another major epigenetic mark, primarily occurs at CpG dinucleotides and is mediated by DNA methyltransferases (DNMTs). In MM, global hypomethylation coexists with site-specific hypermethylation, particularly at tumor suppressor genes, leading to their silencing. This duality contributes to chromosomal instability and loss of regulatory control over key signaling pathways. For example, DNA methylation of the promoter region of UTX was found to be a frequent mechanism of its inactivation in MM [82]. Importantly, histone modifications and DNA methylation are interconnected. Studies have demonstrated that repressive histone marks like H3K27me3 often precede DNA methylation, reinforcing gene silencing in a coordinated manner [84]. This crosstalk forms a stable epigenetic landscape that supports long-term transcriptional repression in malignant plasma cells [85]. These insights into MM epigenetics have also informed therapeutic innovation. Histone deacetylase inhibitors (HDACis) and DNMT inhibitors (DNMTis) are under clinical evaluation for MM, particularly in combination with PIs and immunomodulatory agents. They aim to reverse aberrant silencing of tumor suppressors and resensitize resistant clones to therapy [81].

Epigenetic regulators such as enhancer of zeste homolog 2 (EZH2), histone deacetylases (HDACs), and bromodomain and extra-terminal domain (BET) proteins play central roles in transcriptional control and chromatin organization in MM, contributing to both disease progression and treatment resistance. The histone methyltransferase EZH2 is the catalytic subunit of the polycomb repressive complex 2 (PRC2), which deposits the repressive histone mark H3K27me3. In MM, EZH2 is frequently overexpressed and correlates with poor prognosis by promoting silencing of tumor suppressor genes and dysregulation of cell cycle checkpoints. Inhibition of EZH2 leads to derepression of cyclin-dependent kinase inhibitors and induction of apoptosis in MM cell lines [86]. Moreover, EZH2 inhibitors have demonstrated potent synergy with HDACis such as panobinostat, producing significant anti-myeloma activity and large-scale transcriptomic remodeling in preclinical models [87]. HDACs regulate chromatin condensation and gene silencing by removing acetyl groups from histone tails. Aberrant HDAC activity has been linked to MM pathogenesis by suppressing genes involved in apoptosis and differentiation. Inhibitors of class I and II HDACs, such as panobinostat and ricolinostat, are under clinical investigation or approved in RRMM settings. HDAC inhibition not only induces apoptosis but also disrupts MM-induced suppression of osteoblast differentiation, particularly through reversal of repressive chromatin at key loci like Runx2 [88]. BET proteins, including BRD4, act as “readers” of acetylated histones, recruiting transcriptional machinery to active chromatin regions. BET proteins regulate key oncogenic drivers, such as MYC and BCL2, and are essential for sustaining MM cell proliferation. BET inhibitors like JQ1 disrupt these transcriptional programs and induce cell cycle arrest and apoptosis in MM cells. However, a notable resistance mechanism involves compensatory upregulation of HDAC6, which diminishes BET inhibitor efficacy. Co-inhibition of HDAC6 and BET proteins has demonstrated strong synergy, reducing MYC expression and enhancing MM cell death in both cell lines and patient-derived samples [89]. Combinatorial approaches targeting these pathways simultaneously—e.g., EZH2 + HDAC inhibitors or BET + HDAC6 inhibitors—represent promising therapeutic strategies. These dual-epigenetic blockade regimens enhance cytotoxicity and reprogram malignant transcriptional networks more effectively than monotherapy [90].

MicroRNAs (miRNAs) are short non-coding RNAs (typically 19–25 nucleotides) that regulate gene expression post-transcriptionally by targeting messenger RNAs for degradation or translational inhibition. In MM, numerous miRNAs are dysregulated, functioning as either oncogenes (oncomiRs) or tumor suppressors, thereby influencing cell proliferation, apoptosis, angiogenesis, drug resistance, and interaction with the bone marrow microenvironment [91]. Tumor-suppressive miRNAs, such as miR-29b, miR-34a, and miR-192/194/215, are frequently downregulated in MM [92,93]. For instance, overexpression of miR-29b in MM cell lines induces apoptosis by directly downregulating the anti-apoptotic protein MCL-1, highlighting its therapeutic potential [94]. Conversely, oncogenic miRNAs such as miR-21 and miR-765 are upregulated in MM and promote tumor cell survival, proliferation, and drug resistance. For example, miR-765 exerts its oncogenic function by targeting and downregulating SOX6, a tumor suppressor gene, thereby enhancing MM cell growth [95]. MiRNAs are also intimately connected with other epigenetic mechanisms in MM. MiR-29 family members target DNMTs, and their loss contributes to global DNA hypermethylation. At the same time, c-MYC suppresses miR-29, creating a feedback loop involving c-MYC, miR-29, DNMTs, and miR-34 family, which is silenced by methylation. Disruption of this loop promotes genomic instability and MM progression [96]. Importantly, miRNAs also contribute to drug resistance and relapse. Dysregulated miRNA profiles in extracellular vesicles (EVs) and exosomes influence the tumor microenvironment and can modulate response to therapy. For instance, specific circulating miRNAs (e.g., miR-140-3p, miR-143-3p) have been correlated with high-risk disease and progression to plasma cell leukemia [97]. Due to their regulatory breadth and disease-specific expression patterns, miRNAs have emerged as attractive biomarkers for diagnosis, prognosis, and monitoring of minimal residual disease. Furthermore, miRNA-based therapies, including mimics for tumor-suppressive miRNAs and inhibitors for oncomiRs, are under preclinical and clinical evaluation as adjunctive strategies in MM [98,99,100].

4. Disease Detection and Risk Stratification

4.1. Diagnostic Criteria

The International Myeloma Working Group (IMWG) provides a tiered diagnostic framework for plasma cell disorders, including MGUS, SMM, and MM (

Table 6. IMWG diagnostic criteria for MGUS, SMM, and MM.

Feature	MGUS	SMM	MM
Clonal bone marrow plasma cells (BMPCs)	<10%	≥10% to <60%	≥10% or biopsy-proven plasmacytoma
M-protein in serum (if present)	<3 g/dL	≥3 g/dL (if present) or urinary M-protein ≥500 mg/24 h	Any level (often >3 g/dL)
Myeloma-defining Events	Absent	Absent	Present (≥1 CRAB or SLiM criteria, see below)
CRAB features (myeloma-defining)	None	None	At least one of the following:
			- C: hypercalcemia (Ca >11 mg/dL or >1 mg/dL above normal)
			- R: renal insufficiency (Cr > 2 mg/dL or CrCl < 40 mL/min)
			- A: anemia (Hb < 10 g/dL or >2 g/dL below normal)
			- B: bone lesions (≥1 osteolytic lesion on imaging)
SLiM biomarkers (added in 2014)	Not applicable	Not applicable	Presence of any of the following (even without CRAB):
			- S: ≥60% clonal BMPCs
			- Li: involved/uninvolved serum free light chain (FLC) ratio ≥100
			- M: >1 focal lesion ≥5 mm on MRI
End-organ damage	Absent	Absent	Present or imminent (CRAB or SLiM features)
Progression risk (2-year)	~1%	~10% (standard SMM) to ~50% (high-risk SMM)	100% (active disease requiring treatment)
Recommended management	Observation; no treatment	Observation or clinical trial (if high-risk)	Initiate systemic therapy

). These categories represent a biologic continuum, with escalating risk of progression from asymptomatic to symptomatic malignancy. MGUS is defined by the presence of an M-protein <3 g/dL, <10% clonal plasma cells in the bone marrow, and the absence of myeloma-defining events (MDEs) or end-organ damage attributable to plasma cell dyscrasia. MGUS carries a 1% annual risk of progression to MM or related disorders [1]. SMM represents an intermediate state, characterized by M-protein ≥3 g/dL and/or 10–60% clonal bone marrow plasma cells, with no end-organ damage or biomarker-defined events. Risk of progression is variable, averaging 10% per year for the first five years. Risk models such as the “20/2/20” score (BM plasma cells >20%, M-protein >2 g/dL, FLC ratio >20) help stratify progression risk [101]. Traditionally, the diagnostic criteria for MM were based primarily on clonal bone marrow plasma cells ≥10% or a biopsy-proven plasmacytoma and the presence of symptomatic end-organ damage. This was commonly referred to by the acronym CRAB, which is defined as follows: C: calcium elevation—serum calcium >11 mg/dL or >1 mg/dL above the upper limit of normal; R: renal insufficiency—serum creatinine >2 mg/dL or creatinine clearance <40 mL/min; A: anemia—hemoglobin <10 g/dL or >2 g/dL below the normal limit; B: bone lesions—one or more osteolytic lesions on skeletal radiography, CT, or PET-CT. In 2014, the IMWG expanded the diagnostic criteria for MM by introducing biomarker-based myeloma-defining events, collectively known as SLiM-CRAB: S: ≥60% clonal plasma cells in bone marrow; Li: involved–uninvolved free light chain (FLC) ratio ≥100 (with involved FLC ≥100 mg/L); M: >one focal lesion ≥5 mm on MRI; CRAB: classic criteria (hypercalcemia, renal failure, anemia, bone lesions). Patients meeting one or more SLiM or CRAB criteria are now diagnosed with MM and are candidates for therapy initiation [102]. These changes allow for earlier intervention in high-risk patients—those with an ~80% 2-year risk of progression—even in the absence of symptoms. The incorporation of SLiM markers has led to increased MM diagnoses and greater reliance on advanced imaging (MRI, PET/CT), especially for detecting occult bone lesions [103]. Recent meta-analyses suggest that the prognosis of SLiM-positive MM has improved in the modern era, with median time to progression extending to over 30 months in some cohorts [104].

4.2. Molecular and Genomic Risk Markers

Gene expression profiling (GEP) has emerged as a powerful molecular tool to stratify risk in MM by capturing the transcriptional activity of myeloma cells and identifying biologically distinct subgroups. Unlike cytogenetic tests that identify static genetic abnormalities, GEP reflects dynamic gene regulation, making it especially valuable for assessing tumor behavior and treatment response. The most well-established GEP models include the UAMS 70-gene signature (GEP70), IFM-15, and EMC92, all of which can identify high-risk patients who are more likely to relapse early or exhibit resistance to therapy. For instance, in a cohort receiving lenalidomide and dexamethasone, patients classified as high-risk by GEP70 had a median OS of only 19 months compared to unreached survival in low-risk patients, demonstrating its strong predictive power even with novel agents [105]. GEP also enables classification of MM into transcriptional subtypes that correlate with known chromosomal abnormalities (e.g., t(4;14), t(11;14), 1q gain) and clinical phenotypes. High-risk groups frequently show upregulation of genes related to proliferation (e.g., AURKA), centrosome activity, and MYC signaling [106]. In clinical practice, efforts have been made to integrate GEP data with standard risk markers (e.g., International Staging System [ISS] stages, see

Table 7. Comparison of ISS and R-ISS staging systems (including NGS and FISH integration).

Feature	International Staging System (ISS)	Revised International Staging System (R-ISS)	R2-ISS/NGS-Enhanced or FISH-Enhanced Systems [110]
Year Introduced	2005	2015	Emerging (post-2020), based on genomic insights
Primary Purpose	Estimate prognosis using basic lab values	Refine prognosis by integrating cytogenetics and LDH	Improve risk stratification using NGS and/or advanced FISH data
Parameters Used	- Serum β2-microglobulin - Serum albumin	- ISS components - Chromosomal abnormalities (via FISH) - LDH level	- R-ISS components - TP53 mutation status - 1q gain, del(1p), RAS/BRAF mutations, others
Genetic/cytogenetic input	None	Yes (FISH): del(17p), t(4;14), t(14;16)	Yes (NGS/FISH): includes extended high-risk markers (e.g., 1q gain, biallelic TP53 loss)
LDH incorporation	No	Yes	Yes
Stages defined	- Stage I: β2M < 3.5 mg/L and serum albumin ≥ 3.5 g/dL - Stage II: β2M < 3.5 mg/L; serum albumin < 3.5 g/dL; or β2M 3.5 to 5.5 mg/L, irrespective of serum albumin - Stage III: β2M ≥ 5.5 mg/L	- Stage I: ISS I + no high-risk FISH + normal LDH - Stage II: not R-ISS stage I or III - Stage III: ISS III + high-risk FISH and/or high LDH	- More granular stratification (e.g., 4 groups in R2-ISS) - Assigns scores to each risk marker for cumulative risk
Risk categories	3 tiers (I, II, III)	3 tiers (I, II, III)	Often 4 tiers (e.g., R2-ISS groups I–IV)
Prognostic discrimination	Moderate	Improved over ISS	Superior; captures genetic heterogeneity and clonal complexity
Clinical usefulness	Still used for baseline staging	Widely adopted in clinical trials and risk stratification	Emerging in precision medicine; useful in high-risk MM trials
Limitations	Lacks genetic insight	Does not account for 1q gain, TP53 mutation, or multiple abnormalities	Requires access to NGS and extended FISH; not yet fully standardized
Examples of added markers	N/A	FISH only: del(17p), t(4;14), t(14;16)	NGS/FISH: TP53 mutations, biallelic inactivation, gain(1q21), del(1p), RAS/BRAF mutations, complex karyotype
Progression-free survival (PFS) and OS prediction	Limited precision	Better at stratifying high-risk patients	Most accurate; enables personalized risk-adapted therapy

; cytogenetics) into unified scoring systems such as the HM-metascore, which combines GEP with chromosomal translocations, proliferation index, and druggable gene expression to produce a composite risk assessment [107]. This approach can delineate three prognostic groups with distinct 6-year survival rates of ~89%, 61%, and 19%. Notably, GEP signatures remain predictive after relapse and can track disease evolution. Serial sampling has shown that transcriptional profiles may shift toward higher-risk patterns over time, often reflecting increased proliferation and genomic instability [108]. Therefore, GEP not only informs initial risk assessment but also has potential utility in monitoring progression and treatment adaptation. Although technical barriers and cost have historically limited its widespread adoption, the development of automated tools and open-source platforms is making GEP more accessible in routine settings. These systems now allow clinicians to interpret GEP data alongside conventional factors in real time to guide therapy choices [109].

Next-generation sequencing (NGS) technologies have transformed the landscape of risk assessment in MM, enabling deep interrogation of the mutational spectrum, clonal architecture, and temporal evolution of the disease [2]. Unlike traditional cytogenetics and gene expression profiling, NGS can simultaneously detect point mutations, structural variants, and copy number abnormalities with high resolution, allowing for personalized prognostication and treatment selection. One of the key applications of NGS is in identifying high-risk genetic features that predispose to early relapse and poor survival. Data from the CoMMpass study—a prospective, longitudinal cohort of newly diagnosed MM patients—show that mutations in TP53, IGLL5, and RAS pathway genes, as well as λ-chain translocations, independently predict early progression and reduced OS [111]. This highlights the importance of integrating mutational profiles into baseline risk stratification models. In parallel, efforts have been made to revise existing clinical tools using NGS. A Revised International Staging System based on NGS (R2-ISS) has been proposed (Table 7), which substitutes FISH with NGS-based detection of high-risk cytogenetic abnormalities (e.g., del(17p), t(4;14), t(14;16)). This NGS-enhanced model showed improved sensitivity for high-risk classification and superior discrimination of event-free survival compared to FISH-based R-ISS [112]. NGS also enables longitudinal tracking of clonal evolution, capturing the emergence of new subclones during therapy and relapse. This dynamic view supports adaptive treatment approaches and early identification of resistance mechanisms [113,114,115]. For example, detection of secondary TP53 mutations or gain of 1q during follow-up is often associated with aggressive disease transformation. Another innovative application of NGS is the integration of multi-omic data—including RNA-seq, WES, and WGS—to build composite risk models. The MMNet-286 gene signature, developed from RNA-seq data in the CoMMpass cohort, effectively stratifies patients by risk of early relapse and correlates with genomic instability and MYC pathway activation [116]. Moreover, NGS is being applied to precursor conditions such as SMM to distinguish high-risk progressors. High-risk SMM patients demonstrate increased mutational load, MAPK pathway activation, and frequent copy number alterations, including 1q gain and 8p loss—features detectable via NGS before clinical progression [117].

Next, MRD detection represents a transformative advancement in the monitoring and risk stratification of MM. MRD refers to the small number of malignant plasma cells that persist in a patient after achieving clinical remission, often undetectable by conventional methods like immunofixation or bone marrow morphology. High-sensitivity MRD assays provide powerful prognostic information and are increasingly used to guide therapy and assess depth of response. IMWG defines MRD negativity as the absence of clonal plasma cells at a sensitivity threshold of at least 1 in 100,000 cells (10⁻⁵), achievable via next-generation flow cytometry (NGF) or NGS [118]. NGF-based approaches (e.g., EuroFlow) and NGS platforms (e.g., LymphoSIGHT) have demonstrated comparable sensitivity and are validated for use in clinical trials and routine practice [119]. Achieving MRD negativity is associated with significantly longer PFS and OS, often outperforming traditional response categories such as complete remission (CR). For example, patients who reach sustained MRD negativity over one year have markedly improved long-term outcomes and lower risk of relapse [120]. Importantly, MRD status has proven prognostic value across all treatment phases—including post-induction, post-transplant, and during maintenance therapy [121]. Several techniques are being used for MRD detection, including NGS, NGF, allele-specific oligonucleotide PCR (ASO-PCR), and imaging (PET/CT, MRI) (

Table 8. Comparison of MRD detection techniques in multiple myeloma.

Feature	Next-Generation Sequencing (NGS)	Next-Generation Flow Cytometry (NGF)	Allele-Specific Oligonucleotide PCR (ASO-PCR)	Imaging (PET/CT, MRI)
Technology basis	DNA sequencing of unique IgH rearrangements	Multiparameter flow cytometry with >8–10 color panels	PCR amplification of patient-specific IgH gene rearrangement	Functional/metabolic (PET/CT) and anatomical (MRI) imaging
Sample type	Bone marrow (BM) aspirate	Bone marrow aspirate	Bone marrow aspirate	Whole-body (bone marrow and extramedullary sites)
Sensitivity	Up to 10−6 (1 cell in 1 million)	Up to 10−5 to 10−6	~10−5	~10−2 to 10−3 (lower sensitivity)
Standardization	Highly standardized (e.g., Adaptive ClonoSEQ®)	Standardized by EuroFlow consortium	Requires individualized primers (complex setup)	Variable depending on protocol and scanner
Turnaround time	Moderate (few days to 1–2 weeks)	Fast (24–48 h)	Slow (requires pretreatment sample)	Immediate (PET/CT ~same day; MRI may take longer)
Clonality requirement	Requires identification of clone at diagnosis	Not needed	Requires diagnostic sample for primer design	Not applicable
Applicability	Applicable in ≥90% of patients	Applicable in virtually all patients	Applicable in ~60–70%	Applicable in all patients
Cost and accessibility	Expensive; limited to specialized labs	Moderate cost; increasingly available in tertiary centers	Cost-effective but labor-intensive	Expensive; PET/CT requires radiotracers
Advantages	Highest sensitivity; highly quantitative; standardized	Real-time detection; rapid; does not require diagnostic sample	Highly sensitive; cost-effective for known clone	Detects extramedullary disease and focal lesions missed by BM sampling
Limitations	Requires diagnostic sample; BM only; not real-time	Limited to BM; may miss patchy or extramedullary disease	Technically demanding; not standardized; limited utility	Low sensitivity for minimal disease; can yield false positives due to inflammation
Detects extramedullary disease	No	No	No	Yes
Role in clinical trials	Used in multiple trials as MRD endpoint (e.g., FORTE, GRIFFIN)	Widely used in European MRD studies	Rarely used in modern trials	Used for imaging-based response and disease mapping
Regulatory approval	FDA-cleared (ClonoSEQ for MRD in MM)	CE-marked (EuroFlow)	Not FDA-approved for MRD	Standard clinical use for imaging; not specific for MRD
Integration with IMWG MRD criteria	Yes	Yes	Partially (less common)	Yes (for imaging-defined MRD negativity)

). NGS offers high sensitivity (~10⁻⁶), high reproducibility, and the ability to monitor clonal evolution over time [122], while NGF allows immediate processing and immunophenotypic characterization but requires fresh samples and standardized panels. ASO-PCR has limited use due to patient-specific primer requirements and lower sensitivity than NGS. Imaging could complement marrow-based MRD by identifying extramedullary disease or focal lesions missed in marrow sampling [123]. Emerging innovations include mass spectrometry-based blood tests for MRD, which may reduce the need for invasive bone marrow biopsies and overcome sampling errors from marrow heterogeneity [124]. Despite its clear prognostic value, standardization, cost, and access remain challenges. There is still debate on how to act on MRD results in clinical practice—whether to escalate therapy for MRD-positive patients or de-escalate for those with sustained MRD negativity. Nevertheless, MRD is increasingly incorporated as a surrogate endpoint in clinical trials and is expected to become central in future response-adapted therapeutic strategies [125].

4.3. Imaging Modalities

Modern imaging modalities are central to diagnosing, staging, and monitoring MM, offering critical insight into bone marrow infiltration, skeletal destruction, and extramedullary disease. The three most widely recommended techniques—positron emission tomography–computed tomography (PET/CT), magnetic resonance imaging (MRI), and whole-body low-dose CT (WBLDCT)—each contribute complementary information (

Table 9. Comparison of PET/CT, MRI, and WBLDCT in multiple myeloma imaging.

Feature	PET/CT (FDG-PET/CT)	MRI (Whole-Body or Spine/Pelvis)	WBLDCT (Whole-Body Low-Dose CT)
Imaging modality	Functional (metabolic) + anatomical imaging	Anatomical + functional (tissue contrast, edema detection)	Anatomical (bone structure) only
Radiotracer or contrast	18F-FDG (fluorodeoxyglucose)	None required (but gadolinium can be used)	No contrast or radiotracer required
Radiation exposure	Moderate (higher than WBLDCT)	None	Low-dose radiation
Sensitivity for bone lesions	Moderate to high (especially for active lesions)	Highest for diffuse and focal marrow infiltration	High for osteolytic lesions ≥5 mm
Sensitivity for early disease	Less sensitive for non-metabolically active lesions	Best modality for early bone marrow involvement	Poor at detecting marrow infiltration before bone destruction
Detection of extramedullary disease	Excellent	Limited to soft tissue near bones	Not reliable
Detection of focal lesions	Yes (when metabolically active)	Yes (even before bone destruction)	Yes (if lytic and ≥5 mm)
Detection of diffuse infiltration	Limited	Yes (via signal intensity on STIR/DWI sequences)	No
Detection of healing or inactive lesions	No (FDG-negative)	Yes (but may not distinguish active from inactive)	Yes (can see healed lesions, but not activity)
Use in MRD evaluation	Yes (for imaging-defined MRD negativity)	Yes (e.g., ≥1 focal lesion ≥5 mm = SLiM criterion)	No (not part of IMWG MRD criteria)
Turnaround time	Moderate (requires radiotracer synthesis and uptake time)	Fast (30–60 min; more with contrast)	Fast (10–15 min)
Availability	Moderate (depends on PET facility)	Widely available in tertiary centers	Widely available
Limitations	False negatives in low-FDG-avid MM; false positives from inflammation	Expensive; time-consuming; contraindicated in severe renal failure (if contrast used)	Can miss early or non-lytic disease
Advantages	Identifies active lesions and extramedullary disease	Best for early marrow disease, ideal for SLiM criteria	Best for osteolytic lesion detection; quick and accessible

). PET/CT, particularly with ¹⁸F-FDG, provides both metabolic and anatomic data, making it uniquely effective for detecting active disease, evaluating treatment response, and identifying extramedullary involvement. Studies show PET/CT outperforms skeletal survey in identifying osteolytic lesions and has independent prognostic value both at diagnosis and during remission monitoring [126]. PET/CT can also detect MRD and is recommended for solitary bone plasmacytoma and non-secretory myeloma cases where biochemical markers are insufficient. Meanwhile, MRI, especially whole-body MRI (WB-MRI), is considered the most sensitive modality for detecting diffuse and focal marrow infiltration. It is particularly valuable in smoldering MM and solitary plasmacytoma to reveal occult disease not visible on CT or PET/CT [127]. MRI offers superior soft-tissue contrast and can identify spinal cord compression, vertebral fractures, and subtle marrow changes. Newer sequences like diffusion-weighted imaging (DWI) further enhance sensitivity and MR-based assessments of disease burden [128].

WBLDCT has emerged as a preferred first-line imaging technique, replacing conventional skeletal surveys due to its greater sensitivity for osteolytic lesions, faster acquisition time, and better patient tolerability (Table 9) [129,130]. WBLDCT effectively detects subtle cortical bone changes and is now recommended in the initial staging of newly diagnosed MM. Importantly, radiation dose can be optimized to levels equivalent to or lower than skeletal surveys, maintaining image quality while reducing exposure [131]. Recent comparative studies demonstrate that PET/CT and MRI provide complementary roles, with MRI offering higher sensitivity for marrow infiltration and PET/CT delivering better specificity for active disease and response evaluation [132]. The IMWG now endorses the use of at least one of these cross-sectional modalities (PET/CT, MRI, or WBLDCT) in all patients with suspected MM, with PET/CT preferred for treatment monitoring and MRI for early disease detection [133].

5. Current Therapeutic Strategies

5.1. Frontline Therapy

The treatment paradigm for newly diagnosed MM (NDMM) has evolved significantly with the incorporation of triplet and quadruplet regimens into frontline therapy, aimed at achieving deeper responses, delaying progression, and improving survival. The two foundational triplet combinations—bortezomib, lenalidomide, and dexamethasone (VRd) and daratumumab-containing regimens such as Dara-VRd—are now cornerstones of both transplant-eligible and transplant-ineligible patient care (

Table 10. Frontline pharmacotherapy for newly diagnosed multiple myeloma.

Regimen	Components	Patient Population	Mechanism of Action	Key Clinical Trials	Response Rates/PFS	Special Considerations
VRd	Bortezomib (V) + lenalidomide (R) + dexamethasone (d)	Transplant-eligible and -ineligible	PI + IMiD + steroid	SWOG S0777, IFM 2009	ORR ~90%, CR ~30–40%, median PFS ~43–50 months	Standard of care; neuropathy with bortezomib SC > IV preferred
Dara-VRd	Daratumumab (Dara) + VRd	Transplant-eligible (per FDA/EMA)	Anti-CD38 mAb + PI + IMiD + steroid	GRIFFIN, PERSEUS	sCR ~60%, MRD negativity ≥60%, PFS not reached (superior to VRd)	More infusion time initially; improved depth of response
KRd	Carfilzomib (K) + lenalidomide (R) + dexamethasone (d)	Transplant-eligible and fit patients	2nd-gen PI + IMiD + steroid	FORTE, ENDURANCE	ORR >90%, CR ~40%, MRD negativity ~30–50%	Less neuropathy; more cardiovascular risk than bortezomib
Dara-KRd	Daratumumab + KRd	Transplant-eligible (under investigation)	Anti-CD38 + PI + IMiD + steroid	MASTER, MANHATTAN	MRD negativity >80% in MASTER trial	Deepest responses to date; still investigational in frontline
VTD	Bortezomib + thalidomide + dexamethasone	Transplant-eligible (Europe)	PI + IMiD + steroid	IFM studies	ORR ~85%, CR ~25%, median PFS ~30–35 months	Thalidomide: more neurotoxicity; replaced by lenalidomide in many settings
Dara-VTD	Daratumumab + VTD	Transplant-eligible (Europe)	Anti-CD38 + PI + IMiD + steroid	CASSIOPEIA	sCR 39% vs. 26%; MRD negativity 64% with Dara-VTD	Approved in Europe as standard frontline for transplant-eligible patients
Rd	Lenalidomide + dexamethasone	Transplant-ineligible, frail, or elderly patients	IMiD + steroid	FIRST (MM-020)	PFS ~26 months (continuous Rd)	Better tolerated than triplets; standard for frail or elderly
Dara-Rd	Daratumumab + Rd	Transplant-ineligible (standard of care)	Anti-CD38 + IMiD + steroid	MAIA	ORR ~93%, MRD negativity >30%, PFS not reached (5+ years)	Superior to Rd; preferred triplet for non-transplant-eligible patients
VMP	Bortezomib + melphalan + prednisone	Transplant-ineligible (Europe, elderly)	PI + alkylator + steroid	VISTA	PFS ~24 months	Used mostly in countries without lenalidomide access
Dara-VMP	Daratumumab + VMP	Transplant-ineligible	Anti-CD38 + PI + alkylator + steroid	ALCYONE	sCR ~43%, MRD negativity ~27%, median PFS ~36 months	Approved for elderly/NTI; limited in U.S. use
Ixazomib-Rd	Ixazomib (oral PI) + Rd	Transplant-ineligible, elderly patients	Oral PI + IMiD + steroid	TOURMALINE-MM2 (did not meet PFS endpoint)	Slightly better PFS than Rd; not statistically significant	All-oral regimen; more convenient but less effective than Dara-Rd

PI = proteasome inhibitor; IMiD = immunomodulatory drug; mAb = monoclonal antibody; NTI = non-transplant-eligible; sCR = stringent complete response; MRD = minimal residual disease; PFS = progression-free survival; ORR = overall response rate.

). The VRd regimen has long been a standard of care, especially following the SWOG S0777 trial, which demonstrated superior PFS and OS for VRd compared to lenalidomide and dexamethasone (Rd) alone [134]. However, the addition of daratumumab, an anti-CD38 monoclonal antibody, has redefined expectations for frontline therapy. In the PERSEUS phase III trial, transplant-eligible patients receiving Dara-VRd showed markedly improved outcomes, with 48-month PFS of 84.3% versus 67.7% with VRd alone. Rates of complete response or better and MRD negativity (10⁻⁵) were significantly higher with Dara-VRd (87.9% vs. 70.1% for CR; 75.2% vs. 47.5% for MRD-negative status) [135]. For transplant-ineligible or deferred-transplant candidates, the CEPHEUS study also demonstrated superior efficacy of the quadruplet Dara-VRd over VRd alone. The MRD-negativity rate at 10⁻⁵ was 60.9% with Dara-VRd compared to 39.4% with VRd, and the CR or better rate was 81.2% versus 61.6%, respectively. These deep and sustained responses translated into a 43% reduction in risk of progression or death [136]. In a real-world comparative analysis, daratumumab, lenalidomide, and dexamethasone (DRd) showed improved time-to-next-treatment or death (TTNTD) over VRd in transplant-ineligible patients. The median TTNTD was 37.8 months with DRd compared to 18.7 months with VRd, representing a 42% risk reduction [137]. These findings reinforce the efficacy of daratumumab-containing regimens even in less fit populations. Further supporting the advantage of quadruplet induction therapy, a head-to-head comparison between VRd and daratumumab–bortezomib–thalidomide–dexamethasone (D-VTd) in transplant-eligible patients demonstrated superior response depth with D-VTd. The CR or better rate post-transplant was 90.5% for D-VTd versus 68.6% for VRd [138]. Overall, the accumulating evidence from both clinical trials and real-world data supports the adoption of quadruplet regimens—particularly those including daratumumab—as the new standard of care for newly diagnosed multiple myeloma. These regimens offer higher rates of MRD negativity, longer disease control, and deeper responses across patient subgroups.

Autologous stem cell transplantation (ASCT) remains a cornerstone of treatment in transplant-eligible patients with NDMM despite the emergence of potent induction regimens. ASCT consolidates remission after induction therapy and significantly improves PFS and, in many cases, OS. Evidence from multiple randomized clinical trials and real-world studies confirms that ASCT offers superior outcomes compared to chemotherapy alone. For example, in a real-world Uruguayan cohort of 151 patients treated with high-dose melphalan followed by ASCT, the 36-month OS and PFS were 82.4% and 63.8%, respectively, with a median OS of 98 months—supporting its effectiveness in both high- and low-resource settings [139]. ASCT is particularly important for achieving deep, durable responses. In a large retrospective analysis, 15% of patients who underwent ASCT achieved long-term remission (≥8 years PFS), with higher rates of CR and lower incidence of high-risk features compared to others. This suggests a curative potential in a subset of patients [140]. Importantly, the benefits of ASCT persist in the era of novel agents. The Myeloma XI trial demonstrated that ASCT remains effective even in older patients (≥65), when selected based on fitness rather than age alone. Patients aged 65–75 tolerated ASCT well and derived a significant PFS and OS advantage compared to non-transplanted peers, with hazard ratios of 0.41 (PFS) and 0.51 (OS) [141]. Furthermore, ASCT has shown consistent survival benefit across population-based studies. A California registry analysis of 13,494 NDMM patients found that ASCT recipients had a HR of 0.83 for death, with survival benefits extending across age and treatment cohorts. Notably, delayed ASCT (>12 months after diagnosis) was associated with worse outcomes, reinforcing the value of early transplantation [142]. Current guidelines from IMWG continue to support upfront ASCT as the standard of care for eligible patients. Tandem ASCT may be considered for high-risk cytogenetic subtypes or suboptimal response to the first transplant [143].

5.2. Maintenance and Consolidation Therapy

Lenalidomide is the most widely established maintenance therapy following ASCT in patients with NDMM (

Table 11. Maintenance and consolidation therapies in multiple myeloma.

Therapy Type	Regimen/Drug	Patient Population	Mechanism of Action	Key Trials	Outcomes (PFS/OS)	Special Considerations
Maintenance	Lenalidomide	Post-ASCT; also in non-transplant patients	IMiD; enhances T-cell/NK function, anti-angiogenic	CALGB 100104, IFM 2005-02, Myeloma XI	Median PFS benefit: +18–24 months; OS benefit confirmed	Most widely used; risk of secondary malignancies (SPMs ~7–9%)
Maintenance	Bortezomib (every 2 weeks SC)	High-risk cytogenetics (e.g., del(17p), t(4;14))	Proteasome inhibition → apoptosis	HOVON-65/GMMG-HD4	Improved PFS and OS in high-risk MM	Better tolerability SC vs. IV; neuropathy risk
Maintenance	Ixazomib (oral)	Post-ASCT, especially for convenience	Oral proteasome inhibitor	TOURMALINE-MM3/MM4	PFS improved (~26.5 vs. 21.3 months); no OS benefit yet	All-oral; fewer logistics; mild side effect profile
Maintenance	Daratumumab (monotherapy or with IMiD)	Emerging; studied in high-risk and standard-risk	Anti-CD38 monoclonal antibody	CASSIOPEIA part 2, AURIGA (ongoing)	Early data promising; deepens MRD negativity	SC formulation preferred; infusion reactions mainly in induction
Maintenance	Combination (lenalidomide + bortezomib)	High-risk patients post-ASCT	Dual targeting: IMiD + PI	FORTE, EMN02/HO95	Better MRD negativity; improved PFS in high-risk patients	Considered for double-hit or high-risk cytogenetics
Consolidation	VRd (2–4 cycles post-ASCT)	Transplant-eligible, especially with suboptimal response	PI + IMiD + steroid	IFM/DFCI 2009, EMN02/HO95	Improved depth of response; modest PFS benefit	Useful if no CR/sCR post-ASCT
Consolidation	KRd (carfilzomib-based)	Post-ASCT, high-risk or standard-risk	2nd-gen PI + IMiD + steroid	FORTE (KRd vs. no consolidation)	Higher MRD negativity, PFS advantage	Better tolerated in younger/fit patients
Consolidation	Dara-VTd/Dara-KRd	Under investigation post-ASCT	Anti-CD38-based combinations	GMMG-CONCEPT, MASTER, MANHATTAN	Deep MRD negativity (>80% in MASTER)	Still investigational; tailored to MRD-guided therapy
Maintenance (SMM)	Lenalidomide ± dexamethasone	High-risk smoldering MM	Delays progression to active MM	ECOG E3A06, ASCENT-SMM	3-year PFS ~90% vs. ~60% (control)	For selected high-risk SMM patients under clinical trial protocols

Maintenance therapy: low-dose, long-term treatment post-induction/ASCT to prolong remission; consolidation therapy: short-term, intensive treatment post-ASCT to deepen response; ASCT = autologous stem cell transplant; PFS = progression-free survival; OS = overall survival; MRD = minimal residual disease; PI = proteasome inhibitor; IMiD = immunomodulatory drug.

). It functions through direct tumoricidal activity, inhibition of angiogenesis, and immune modulation, and it has consistently been shown to prolong PFS and OS. Two pivotal phase III trials—IFM 2005-02 and CALGB 100104—first established the role of lenalidomide in this setting [144,145,146]. Both demonstrated significant PFS improvements, and CALGB 100104 also showed a survival benefit. A meta-analysis pooling these and the GIMEMA trial data confirmed a statistically significant OS benefit from lenalidomide maintenance post-ASCT [147]. Real-world data from the Connect MM registry further validated these findings. Patients receiving lenalidomide-only maintenance had significantly longer PFS (54.5 vs. 30.4 months) and OS (HR 0.45; p = 0.001) compared to those receiving no maintenance. The benefit was seen without a significant increase in healthcare resource utilization, supporting lenalidomide’s feasibility outside clinical trial settings [148]. Lenalidomide also appears to enhance depth of response post-transplant. In a cohort of 139 patients, 38.1% achieved maximal response during maintenance, and 34.3% of initially MRD-positive patients converted to MRD negativity, which correlated with prolonged PFS [149]. These findings support the concept that maintenance therapy not only sustains remission but can deepen response over time, especially in patients with residual disease. The Myeloma XI phase III trial also reinforced lenalidomide’s superiority over observation alone, with median PFS of 39 months vs. 20 months. Importantly, subgroup analyses showed that both standard-risk and high-risk cytogenetic patients benefited, although the OS advantage was more evident in transplant-eligible patients [150]. Concerns regarding toxicity and second primary malignancies (SPMs) have been noted, particularly hematologic adverse events. However, the risk–benefit ratio remains favorable, as the survival benefit generally outweighs the risk of SPMs [147]. Furthermore, studies suggest that lower starting doses (e.g., 5–10 mg/day) are feasible and tolerable in older patients or those with comorbidities [151].

PIs, particularly bortezomib and ixazomib, have emerged as important maintenance options after ASCT, especially in high-risk MM (Table 11). While lenalidomide remains the standard maintenance therapy for most patients, PI-based maintenance is often preferred in those with adverse cytogenetic features or intolerance to immunomodulatory drugs (IMiDs). A systematic review and meta-analysis evaluating bortezomib and ixazomib maintenance post-ASCT found that PI-based strategies significantly improved PFS (HR 0.75, 95% CI 0.67–0.85) and deepened hematologic responses compared to placebo or thalidomide, without increasing SPM risk [152]. While OS data were less conclusive, these results support the efficacy of PIs in prolonging disease control. Real-world and retrospective data further support the preferential use of PIs in high-risk cytogenetic subgroups, such as patients with del(17p), t(4;14), or gain(1q). A Mayo Clinic cohort showed that high-risk patients on PI-based maintenance had better PFS compared to those on IMiD-only or no maintenance, with the benefit being most pronounced in those who underwent early transplant (<12 months from diagnosis). In this group, bortezomib-containing maintenance improved long-term outcomes and narrowed the survival gap between high-risk and standard-risk patients [153]. In the HOVON-65/GMMG-HD4 trial, bortezomib maintenance for two years post-ASCT showed clear PFS and OS advantages over thalidomide, particularly among patients with high-risk cytogenetics. Notably, the trial also demonstrated a reduction in renal failure, further supporting PI use in renally impaired MM patients [154,155]. Among 827 patients, those receiving bortezomib-based induction (PAD: bortezomib, doxorubicin, dexamethasone) followed by bortezomib maintenance showed significantly higher complete response rates (31% vs. 15% after induction; 49% vs. 34% during maintenance) compared to the VAD (vincristine-based) regimen. After a median follow-up of 41 months, the PAD group had improved PFS (35 vs. 28 months; HR 0.75, p = 0.002) and OS (HR 0.77, p = 0.049). Notably, high-risk subgroups—including patients with elevated creatinine (>2 mg/dL) and those with del(17p13)—experienced substantial survival benefits from bortezomib. For instance, in patients with renal impairment, bortezomib extended PFS from 13 to 30 months and OS from 21 to 54 months [156]. After a median follow-up of 96 months, bortezomib-based treatment still significantly improved PFS (HR = 0.76, p = 0.001), though OS was not statistically different between arms (HR = 0.89, p = 0.24). Importantly, the adverse impact of high-risk features such as deletion 17p13 and renal impairment was mitigated in the PAD arm but persisted in the VAD group. Rates of SPM were similar across arms (7%, p = 0.73), and OS after first relapse was comparable (HR = 1.02, p = 0.85). The study confirms that bortezomib during induction and maintenance provides sustained PFS benefit in MM without increasing long-term toxicity risk [157].

More recently, ixazomib, an oral PI, has shown promise in the TOURMALINE-MM3 and MM4 trials (Table 11). In the TOURMALINE-MM4 study involving transplant-ineligible patients, ixazomib maintenance significantly improved PFS (17.4 vs. 9.4 months) compared to placebo, with a 34% risk reduction in disease progression or death and a favorable safety profile [158]. These findings suggest that ixazomib may be a practical option for patients seeking an oral, less toxic maintenance alternative. Despite these benefits, some limitations remain. Toxicity—particularly peripheral neuropathy with bortezomib—can affect adherence, although weekly dosing and subcutaneous administration have mitigated these effects. Additionally, head-to-head comparisons between lenalidomide and PIs for maintenance are lacking, and ongoing trials aim to clarify optimal maintenance selection and duration.

5.3. Management of High-Risk Disease

High-risk multiple myeloma (HRMM) is defined by the presence of specific cytogenetic abnormalities associated with poor clinical outcomes. Among these, TP53 alterations, especially double-hit MM (coexistence of TP53 deletion and mutation), represent some of the most aggressive and treatment-resistant forms of the disease. A landmark study using SNP array and next-generation sequencing found that biallelic TP53 inactivation (double-hit) conferred the worst clinical outcomes, with significantly shorter PFS and OS compared to single-hit or TP53 wild-type patients. Double-hit MM patients had a PFS HR of 3.34 and an OS HR of 3.47, underscoring their aggressive biology [159]. Another large-scale study confirmed that del(17p) alone also confers poor prognosis, but the combination with TP53 mutation defines the most aggressive subgroup—double-hit MM—with median OS of just 36 months [160]. Ultra-deep sequencing techniques have revealed that TP53 mutations are more prevalent than previously recognized, particularly at relapse, and double-hit MM is associated with younger age at diagnosis, higher marrow plasma cell infiltration, and a higher rate of extramedullary disease [161]. From a therapeutic standpoint, TP53 inactivation confers resistance to alkylating agents like melphalan and doxorubicin, while PIs retain efficacy, even in TP53-deficient models [162]. This observation supports the use of PI-based maintenance (e.g., bortezomib) in patients with del(17p) or TP53 mutation. Double-hit MM is also associated with higher clonal fitness and therapy resistance. Functional in vitro models show that both monoallelic and biallelic TP53 lesions increase resistance to genotoxic agents and enhance cell fitness, especially under drug selection pressure [163]. These findings highlight the urgent need for personalized approaches, including early ASCT and PI-based consolidation in frontline therapy, triplet/quadruplet regimens incorporating daratumumab, and participation in clinical trials evaluating CAR-T, bispecific antibodies, or TP53-reactivating agents. Despite aggressive therapy, outcomes remain suboptimal. A real-world study of double- and triple-hit MM reported median OS of just 13 months, with early deaths from progressive disease common, emphasizing the unmet need in this population [164].

Thus, the management of HRMM—characterized by adverse cytogenetic profiles like del(17p), t(4;14), t(14;16), gain(1q), and TP53 mutations—requires an individualized and intensified therapeutic approach due to the aggressive nature and poor prognosis of the disease (

Table 12. Comparison of management strategies in standard-risk vs. high-risk multiple myeloma.

Aspect	Standard-Risk MM	High-Risk MM
Definition	Absence of high-risk cytogenetic features or clinical markers	Presence of high-risk cytogenetics (e.g., del(17p), t(4;14), t(14;16), gain 1q), or double-hit MM (e.g., biallelic TP53)
Common cytogenetic features	Trisomies (hyperdiploid), t(11;14), no 1q gain	del(17p), t(4;14), t(14;16), gain(1q21), del(1p), complex karyotype, TP53 mutation
Initial risk assessment tools	FISH, R-ISS	R2-ISS, NGS, GEP, FISH
Frontline induction therapy	Triplet regimens (e.g., VRd, Dara-VRd)	Intensified regimens (e.g., KRd, Dara-KRd, quadruplets)
Stem cell transplant	Single ASCT	Early ASCT recommended; tandem ASCT considered in some high-risk cases
Consolidation therapy	Optional; may use short course of VRd post-ASCT	Recommended (e.g., KRd, Dara-KRd) to deepen response
Maintenance therapy	Lenalidomide monotherapy	Bortezomib-based (e.g., bortezomib ± lenalidomide); consider dual-agent maintenance
MRD monitoring	MRD monitoring used, but not always guiding therapy	Essential; MRD status guides continuation or intensification
Treatment goals	Durable remission; manageable toxicity	Achieve and sustain MRD negativity; delay clonal evolution
Role of clinical trials	Optional; used for access to novel agents	Strongly encouraged to access cutting-edge immunotherapy/targeted regimens
Preferred immunotherapy options	Anti-CD38 mAbs (e.g., daratumumab) in standard regimens	Early use of immunotherapy (e.g., Dara, BiTEs, CAR-T in trials)
Relapse management	Sequential therapy (PI/IMiD/monoclonal)	Early use of novel agents (e.g., selinexor, CAR-T, bispecifics)
Prognosis	Median PFS > 5 years; OS > 8–10 years	Median PFS ~2–3 years; OS ~4–6 years (may vary by subtype)
Monitoring frequency	Routine labs every 1–3 months	More frequent labs, imaging, and MRD assessments
Emerging approaches	MRD-guided de-escalation (investigational)	MRD-guided intensification, genomics-based precision therapy

). Traditional uniform strategies are no longer adequate, and mounting evidence supports risk-adapted regimens designed to overcome resistance and deepen remission. Recent pooled analyses of phase III trials have demonstrated that continuous therapy incorporating PIs and IMiDs significantly improves outcomes in high-risk patients. Specifically, high-risk patients treated with continuous bortezomib-IMiD maintenance had greater PFS and OS compared to those receiving fixed-duration therapy, effectively narrowing the survival gap with standard-risk patients [165]. Quadruplet induction regimens, such as Dara-VRd, are strongly recommended for HRMM due to their ability to achieve deeper and more durable responses, including MRD negativity. In transplant-eligible patients, this approach enhances MRD-negative rates and has been associated with longer PFS compared to triplet regimens [166]. In the post-transplant setting, double or tandem ASCT is a viable strategy for patients with high-risk features. Retrospective analyses show that tandem transplants can lead to deeper responses and may improve survival outcomes, especially in patients with t(4;14) or del(17p), although this benefit is not uniform and needs further validation through prospective trials [167]. Maintenance therapy also needs to be intensified. While lenalidomide is the standard for standard-risk MM, high-risk patients benefit more from combination maintenance with a PI (e.g., bortezomib or carfilzomib) plus lenalidomide, which has shown improved depth of response and survival in high-risk cohorts [168]. Moreover, risk-adapted therapy must be dynamic, guided by MRD status and evolving molecular markers. MRD negativity, sustained over time, is emerging as the most powerful prognostic tool and may eventually guide de-escalation or intensification of therapy [167]. To summarize, tailored intensive strategies for HRMM integrate quadruplet induction, early transplant (often tandem), and dual-agent maintenance, with therapy modulation based on MRD status. These approaches are essential for improving outcomes in a subgroup traditionally associated with rapid relapse and inferior survival.

5.4. Management of Bone Loss in Multiple Myeloma

Bisphosphonates and denosumab are cornerstone agents in the management of bone disease in MM. Both classes of drugs act by inhibiting osteoclast-mediated bone resorption, though through distinct mechanisms—bisphosphonates incorporate into the bone matrix and disrupt osteoclast function upon resorption, while denosumab is a monoclonal antibody that neutralizes RANKL, a key driver of osteoclastogenesis. Bisphosphonates, particularly zoledronic acid and clodronate, have long been used to reduce skeletal-related events (SREs) and bone pain in MM. Zoledronic acid has been shown to significantly reduce the risk of SREs and may also improve OS, although it is associated with nephrotoxicity and a risk of osteonecrosis of the jaw (ONJ) with long-term use [169]. ONJ remains a major concern, occurring in 6–9% of patients, particularly with prolonged therapy. Preventative dental assessments are essential before and during treatment [170]. Meanwhile, denosumab has emerged as a potent alternative, particularly in patients with renal dysfunction. In the pivotal phase III study comparing denosumab with zoledronic acid, denosumab was non-inferior in delaying time to first SRE and demonstrated a more favorable renal safety profile [171]. A systematic review concluded that denosumab offers comparable efficacy with potentially lower renal toxicity, although both agents have similar rates of ONJ [169]. Denosumab is administered subcutaneously and does not require renal dose adjustments, enhancing its convenience in clinical settings. Both drugs are also used in the treatment of hypercalcemia of malignancy, with denosumab being effective even in patients refractory to bisphosphonates [171]. However, hypocalcemia is more common with denosumab, especially in patients with impaired renal function, necessitating calcium and vitamin D supplementation [172].

6. Immunotherapy and Targeted Agents

6.1. Monoclonal Antibodies

Targeting CD38, a transmembrane glycoprotein highly expressed on malignant plasma cells, has revolutionized MM therapy. Two anti-CD38 monoclonal antibodies—daratumumab and isatuximab—are now widely used across treatment lines, from newly diagnosed to RRMM, due to their deep and durable anti-myeloma effects (

Table 13. Comparison of anti-CD38 and anti-SLAMF7 monoclonal antibodies in multiple myeloma management.

Feature	Daratumumab	Isatuximab	Elotuzumab
Target Antigen	CD38	CD38	SLAMF7 (CS1)
Type of Antibody	Fully human IgG1κ	Chimeric IgG1	Humanized IgG1
Mechanism of Action	- Direct apoptosis (crosslinking) - CDC, ADCC, ADCP - Immune modulation (Treg depletion)	- Direct apoptosis (even without crosslinking) - ADCC, CDC, ADCP - Inhibits CD38 ectoenzyme	- Enhances NK cell activation via SLAMF7 - ADCC (no direct cytotoxicity)
FDA Approval (MM)	Yes (monotherapy and in combinations, frontline and relapsed)	Yes (in combination regimens for relapsed MM)	Yes (only in combination with IMiDs for relapsed MM)
Common Combinations	- Dara-Rd (MAIA) - Dara-VRd (GRIFFIN) - Dara-KRd - Dara-Pd (APOLLO)	- Isa-Rd (IKEMA) - Isa-Kd (IKEMA)	- Elo-Rd (ELOQUENT-2) - Elo-Pd (ELOQUENT-3)
Line of Use	- Frontline (NTI and transplant-eligible) - Relapsed/refractory (RRMM)	- RRMM (second line and beyond)	- RRMM (second line and beyond, not approved for frontline)
Route of Administration	IV and SC (subcutaneous)	IV only	IV only
Infusion Time	IV: ~3–6 h; SC: ~5 min	IV: ~3 h	IV: ~2–3 h
Infusion Reactions	~40% (mostly first dose); lower with SC	~35% first-dose reactions	~10–15% (generally mild)
Effect on MRD	High MRD negativity when combined with IMiDs/PI	Comparable MRD negativity in IKEMA study	Not MRD-focused; less depth of response
Efficacy in High-Risk MM	Active; used in high-risk combinations	Active; included in some high-risk trials	Less data in high-risk subsets
Unique Features	- SC formulation reduces infusion burden - Depletes Tregs and Bregs	- Inhibits CD38 enzymatic function - More potent direct killing	- Synergizes with NK cells; no single-agent activity
Adverse Effects	Cytopenias, infections, IRRs, fatigue	Similar to daratumumab	Fatigue, diarrhea, cough, IRRs
Use with IMiDs or PIs	Works well with both IMiDs and PIs	Primarily used with IMiDs	Only approved with IMiDs (lenalidomide, pomalidomide)
Trial Highlights	- MAIA (Dara-Rd) - GRIFFIN (Dara-VRd) - CASSIOPEIA (Dara-VTd)	- IKEMA (Isa-Kd) - ICARIA-MM (Isa-Pd)	- ELOQUENT-2 (Elo-Rd) - ELOQUENT-3 (Elo-Pd)

). Daratumumab is a fully human IgG1κ monoclonal antibody that induces myeloma cell death through multiple mechanisms, including complement-dependent cytotoxicity (CDC), antibody-dependent cellular cytotoxicity (ADCC), phagocytosis, and direct apoptosis. In addition, it depletes CD38+ immunosuppressive cells, restoring immune effector function [173]. Daratumumab is approved for both transplant-eligible and ineligible newly diagnosed MM patients in combination regimens such as Dara-VRd, which has shown high rates of MRD negativity and prolonged PFS [135]. Isatuximab, while also targeting CD38, binds to a distinct, non-overlapping epitope compared to daratumumab. It exhibits unique properties, including direct pro-apoptotic activity without crosslinking and less reliance on complement activation, making it effective even in low-CD38-expressing tumors [174]. Isatuximab is approved in combination with pomalidomide and dexamethasone (Isa-Pd) or carfilzomib and dexamethasone (Isa-Kd) for patients with RRMM, including those with high-risk features. Real-world evidence supports the continued efficacy of anti-CD38 therapy across multiple lines of treatment, even after prior exposure. In a cohort of relapsed MM patients receiving more than one anti-CD38 regimen, the overall response rate remained stable (~40%) in the first two exposures, suggesting clinical benefit from retreatment strategies [175]. Despite their success, resistance and CD38 downregulation can occur with repeated exposure. Recent studies report biallelic loss of CD38 in a subset of patients post-treatment, conferring resistance to both agents. Interestingly, some CD38 mutations selectively affect daratumumab binding but retain sensitivity to isatuximab, suggesting a potential role for sequential anti-CD38 therapy guided by molecular profiling [176]. Combination approaches to enhance anti-CD38 efficacy are under investigation. Inhibitors of CD46 and CD59, which block CDC, have been shown to increase the cytotoxicity of daratumumab and isatuximab in vitro and in vivo [177].

Next is elotuzumab, a first-in-class, humanized IgG1 monoclonal antibody that targets Signaling Lymphocytic Activation Molecule Family member 7 (SLAMF7), a glycoprotein highly and selectively expressed on both normal and malignant plasma cells as well as NK cells (Table 13). Unlike other monoclonal antibodies, elotuzumab exhibits little direct cytotoxicity as a monotherapy but shows potent activity when combined with IMiDs through immune-cell-mediated mechanisms. Elotuzumab exerts its anti-myeloma effects primarily by enhancing NK cell-mediated ADCC. It binds SLAMF7 on MM cells and simultaneously activates NK cells via CD16, thereby bridging immune effectors with tumor targets [178]. Additionally, elotuzumab can provide costimulatory signals to NK cells via direct SLAMF7 engagement, further enhancing their cytotoxic potential even in CD16-independent pathways [179]. The clinical efficacy of elotuzumab was demonstrated in the ELOQUENT-2 trial, which led to its FDA approval in 2015. In this phase III trial, patients with RRMM who received elotuzumab in combination with lenalidomide and dexamethasone (ELd) had a 30% reduction in the risk of progression or death compared to those receiving lenalidomide and dexamethasone alone [180]. Follow-up studies confirmed the durable benefit and favorable safety profile of ELd in various patient populations, including those with high-risk cytogenetics like t(4;14) and del(17p) [181]. Preclinical and clinical studies also show that elotuzumab synergizes with IMiDs like lenalidomide by promoting NK cell activation through cytokine production (e.g., IL-2, TNF-α) and upregulation of adhesion molecules [182]. This synergy underlines its limited single-agent activity but robust efficacy in combination regimens. While elotuzumab is less frequently used in the frontline setting due to the dominance of anti-CD38 antibodies, it retains value in RRMM, especially for patients previously exposed to daratumumab. Importantly, SLAMF7 expression is retained even in advanced or extramedullary disease, supporting its continued utility in resistant cases [183].

6.2. Antibody–Drug Conjugates

Belantamab mafodotin is a first-in-class antibody–drug conjugate (ADC) targeting B-cell maturation antigen (BCMA), a transmembrane protein highly expressed on malignant plasma cells. It comprises an afucosylated anti-BCMA monoclonal antibody conjugated to the cytotoxic agent monomethyl auristatin F (MMAF). Upon binding to BCMA, the drug is internalized and delivers MMAF directly into the cell, disrupting microtubule dynamics and inducing apoptosis [184]. Belantamab mafodotin was granted accelerated FDA approval in 2020 for patients with RRMM who had received at least four prior therapies, including a proteasome inhibitor, an immunomodulatory drug, and an anti-CD38 antibody. The approval was based on results from the DREAMM-2 trial, which demonstrated a 32–35% overall response rate (ORR) and a median duration of response of 6.2–12.5 months in heavily pretreated patients [185,186,187]. Notably, responses occurred across high-risk cytogenetic subgroups, including patients with extramedullary disease. Despite its clinical activity, ocular toxicity—particularly keratopathy—is a major limiting factor [187]. Up to 69% of patients develop corneal microcyst-like epithelial changes (MECs), often leading to visual disturbances, dose delays, or discontinuation [188]. This adverse event profile prompted the withdrawal of belantamab mafodotin from the U.S. market in 2022 pending confirmatory trial outcomes, though it remains available under compassionate use and in clinical studies.

Ongoing trials are exploring combination strategies to enhance efficacy and reduce toxicity. In the BELARD study, combining belantamab with lenalidomide and dexamethasone (belantamab-Rd) in transplant-ineligible NDMM patients yielded encouraging response rates and a manageable safety profile, suggesting potential synergy with ImiDs [189]. Likewise, in the BELACARD study, belantamab combined with carfilzomib and dexamethasone showed promising efficacy in RRMM, supporting continued development in multidrug regimens [190]. Furthermore, gamma-secretase inhibitors (GSIs; e.g., nirogacestat) that prevent BCMA shedding have demonstrated synergistic activity with belantamab mafodotin in preclinical models, leading to enhanced ADCC and tumor cell killing—even in low-BCMA-expressing cells [191]. This approach is being tested in the ongoing DREAMM-5 [192].

6.3. CAR-T-Cell Therapy

Chimeric antigen receptor T-cell (CAR-T) therapy targeting BCMA has transformed the treatment landscape for RRMM. BCMA is selectively expressed on malignant plasma cells, making it an ideal target for cellular immunotherapy. Two FDA-approved BCMA-targeting CAR-T therapies—idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel)—have demonstrated unprecedented efficacy in heavily pretreated patients (

Table 14. Comparison of idecabtagene vicleucel (ide-cel) vs. ciltacabtagene autoleucel (cilta-cel) in multiple myeloma.

Feature	Idecabtagene Vicleucel (ide-cel; Abecma®)	Ciltacabtagene Autoleucel (cilta-cel; Carvykti®)
Target antigen	B-cell maturation antigen (BCMA)	BCMA
CAR structure	Single-chain variable fragment (scFv)	Two BCMA-binding domains (dual-epitope targeting scFvs)
CAR-T-cell generation	Lentiviral vector, autologous T cells	Lentiviral vector, autologous T cells
FDA approval	March 2021	February 2022
Indication	Triple-class refractory MM (≥4 prior lines)	Triple-class refractory MM (≥4 prior lines)
Key clinical trial	KarMMa (Phase 2)	CARTITUDE-1 (Phase 1b/2)
Overall response rate (ORR)	~73%	~98%
Stringent complete response (sCR)	~33%	~80%
Median progression-free survival	~8.8 months (median follow-up 13.3 months)	Not reached at 27 months; PFS at 2 years: ~61%
Median overall survival	~24.8 months	Not reached at 2 years; OS ~74%
Time to response	~1 month	~1 month
Time from apheresis to infusion	~4–5 weeks	~4–6 weeks
Toxicity—CRS (cytokine release syndrome)	~84% (mostly Grade 1–2); Grade ≥3 in ~5%	~95%; mostly Grade 1–2; Grade ≥3 in ~4%
Toxicity—ICANS (neurotoxicity)	~18% (Grade ≥3: 3%)	~21% (Grade ≥3: 10%)
Unique features	First FDA-approved BCMA CAR-T; robust data in heavily pretreated patients	Dual-epitope binding = higher avidity; very deep and durable responses
Dosing regimen	Single infusion of 150–450 × 106 CAR+ T cells	Single infusion of 0.75 × 106 CAR+ T cells/kg
Emerging directions	Used in KarMMa-3 (earlier lines); potential combinations	CARTITUDE-4: early relapse; being explored for frontline high-risk MM

). Ide-cel was the first BCMA CAR-T product to receive regulatory approval. In the pivotal KarMMa trial, ide-cel achieved an ORR of 73% and median PFS of 8.8 months, with durable responses in patients with a median of six prior therapies [193]. While ide-cel showed manageable rates of cytokine release syndrome (CRS) and neurotoxicity, long-term remissions were limited in many patients due to CAR-T-cell exhaustion and antigen escape. Meanwhile, cilta-cel, a second-generation CAR-T with two BCMA-binding domains, has demonstrated deeper and more durable responses. In the CARTITUDE-1 study, cilta-cel yielded a 98% ORR, with 83% achieving stringent CR (sCR) and median PFS exceeding 34.9 months, setting a new benchmark in RRMM [194]. Subsequent trials, such as CARTITUDE-4, have shown efficacy even earlier in treatment lines, suggesting potential for curative intent [195]. Despite the promise of BCMA CAR-T, relapse remains a concern, occurring in 27–64% of responders. Strategies to improve outcomes include retreatment with alternative BCMA CAR-T constructs and combination with bispecific antibodies targeting other antigens (e.g., FcRH5) [196]. Cilta-cel retreatment has shown feasibility, with 80% response rates in patients previously exposed to ide-cel.

Both ide-cel and cilta-cel are associated with immune-mediated toxicities, particularly CRS and immune effector cell-associated neurotoxicity syndrome (ICANS) (Table 14). These toxicities result from excessive immune activation and inflammation following T-cell engagement and expansion and require vigilant monitoring and prompt management to ensure patient safety. CRS is the most common adverse event following CAR-T therapy, occurring in up to 90% of patients, but is mostly Grade 1–2. It is triggered by massive cytokine release from activated CAR-T cells and myeloid cells, leading to fever, hypotension, hypoxia, and in severe cases, multi-organ dysfunction. The standard treatment for moderate to severe CRS (Grade ≥2) includes tocilizumab, an IL-6 receptor antagonist, and corticosteroids for more severe or refractory cases [197]. Early use of these agents has been shown to reduce toxicity without impairing CAR-T efficacy [198]. In contrast, ICANS is a distinct neurotoxicity syndrome often following CRS, characterized by symptoms ranging from mild confusion and aphasia to seizures and cerebral edema. The pathophysiology involves disruption of the blood–brain barrier and neuroinflammation driven by systemic cytokines. Management of ICANS includes high-dose corticosteroids (e.g., dexamethasone), and unlike CRS, tocilizumab is not effective due to limited blood–brain barrier penetration [199]. For patients with steroid-refractory ICANS, other agents such as anakinra (IL-1R antagonist) and anti-TNF therapies are under investigation [200]. Neurotoxicities, including Parkinsonism-like syndromes and cranial nerve palsies, are rare but more frequently reported with cilta-cel due to its higher expansion kinetics [201]. High peak CAR-T expansion is associated with both improved response and increased neurotoxicity risk [202]. Infections are also a notable complication, with hypogammaglobulinemia and prolonged cytopenias contributing to bacterial and viral infections. However, comparative real-world analyses show similar infection risks between ide-cel and cilta-cel, reinforcing the need for infection prophylaxis and monitoring [203].

The American Society for Transplantation and Cellular Therapy (ASTCT) provides widely accepted grading criteria for CRS (

Table 15. Grading of cytokine release syndrome (CRS).

Grade	Fever ≥ 38 °C	Hypotension	Hypoxia
Grade 1	Present	None	None
Grade 2	Present	Does not require vasopressors; responds to fluids	Requires low-flow oxygen (nasal cannula ≤6 L/min)
Grade 3	Present	Requires vasopressors or multiple boluses	Requires high-flow oxygen, CPAP, or BiPAP
Grade 4	Present	Requires multiple vasopressors ± ventilatory support	Requires mechanical ventilation

) and ICANS (

Table 16. Grading of immune effector cell-associated neurotoxicity syndrome (ICANS).

Grade	ICE Score	Level of Consciousness	Seizure	Motor Findings	Cerebral Edema
Grade 1	7–9	Awaken spontaneously	None	None	None
Grade 2	3–6	Awaken to voice	None	None	None
Grade 3	0–2	Arousable with tactile stimulation	Any seizure	None	Focal/local edema in neuroimaging
Grade 4	0	Unarousable or obtunded; stupor or coma	Life-threatening prolonged seizure	Deep focal motor weakness, such as hemiparesis or paraparesis	Diffuse cerebral edema in neuroimaging; decerebrate or decorticate

ICE score components (maximum 10 points): orientation to year, month, city, hospital = 4; naming 3 objects = 3; following simple command = 1; writing a standard sentence = 1; attention (counting backwards from 100 by 10) = 1.

), guiding when to escalate therapy [204]. For example, Grade 1 CRS is managed supportively, while Grade 2 or higher typically warrants tocilizumab and possibly corticosteroids. ICANS is graded based on the Immune Effector Cell-Associated Encephalopathy (ICE) score and neurological findings, with steroids initiated for Grade ≥2 [205]. Timing and duration of these toxicities are critical for post-infusion monitoring. A multicenter study found that late-onset CRS or ICANS beyond day 14 is rare, prompting discussion of shortening the standard 4-week monitoring period to reduce burden on patients and healthcare systems [206]. Importantly, while CRS and ICANS are not entirely avoidable, recent evidence suggests that effective toxicity management does not compromise efficacy. In fact, early intervention may even improve response rates by minimizing immune exhaustion [207].

The resistance to CAR T-cell therapy in MM is driven by a complex interplay of host-related factors, tumor-intrinsic mechanisms, the immunosuppressive tumor microenvironment, and characteristics of the CAR T-cell product itself. While traditional prognostic factors like age are less predictive in CAR T-cell therapy, physical fitness and performance status remain critical, with conditions like sarcopenia and cachexia linked to poor outcomes. Intriguingly, visceral adiposity has a paradoxical role—associated with more severe CRS due to increased IL-6 production from adipose tissue but also with better PFS and OS, possibly through “metaflammation,” a pro-inflammatory transformation of fat tissue that supports immune activation. Additionally, renal function and weight significantly affect fludarabine pharmacokinetics, influencing lymphodepletion efficiency. Notably, broad-spectrum antibiotics prior to infusion were associated with reduced CAR T efficacy and increased neurotoxicity, suggesting the gut microbiome’s critical role in modulating CAR T-cell response [208]. Meanwhile, a key tumor-intrinsic resistance mechanism is antigen loss or downregulation, particularly BCMA. Although biallelic BCMA deletions are rare, partial or transient downregulation without genetic mutations can result in antigen-low escape. This may reflect clonal selection under continuous immunologic pressure. Notably, responses to CAR T-cell therapy have still been observed in patients with prior anti-BCMA therapy exposure, suggesting some capacity to overcome BCMA-low clones. Additional resistance mechanisms include upregulation of inhibitory ligands (e.g., PD-L1), loss of costimulatory molecules (e.g., CD58), and altered apoptotic machinery, which together impair CAR T-cell cytotoxicity even when the target antigen is present [208]. The bone marrow tumor microenvironment in MM also acts as a formidable barrier to CAR T-cell efficacy. It harbors suppressive immune cells such as M2 macrophages, Tregs, and MDSCs, all of which foster CAR T-cell exhaustion and reduce infiltration. For example, STAT3-activated tumor-associated macrophages promote plasma cell growth and resistance to immune killing. Additionally, extracellular matrix components and abnormal vasculature physically restrict CAR T-cell trafficking. Immune profiling studies have also shown that pre-existing inflammatory signatures and serum markers like CRP, IL-6, and ferritin correlate with poor CAR T expansion and response, emphasizing the predictive value of systemic inflammation [208].

Next, the phenotype and fitness of T-cells used in manufacturing also critically affect outcomes. Products enriched for less differentiated (naïve or early memory) T cells, especially CCR7+CD45RA+ CD8+ T-cells, exhibit better expansion and persistence, translating into superior efficacy. Conversely, T-cell exhaustion, driven by chronic antigen stimulation and tonic CAR signaling, leads to functional decline, marked by upregulation of PD-1, LAG-3, TIM-3, and TIGIT. This is accompanied by epigenetic reprogramming and loss of cytokine production. Moreover, some patients exhibit CAR-expressing regulatory T cells (CAR-Tregs) post-infusion, which can suppress overall CAR T expansion and function, further compromising response [208]. Finally, a critical but underappreciated barrier to durable CAR T efficacy is immune rejection of the CAR T-cells themselves. Most CAR constructs use murine-derived single-chain variable fragments (scFvs), which are immunogenic. Patients may develop anti-CAR antibodies or CAR-specific cytotoxic T cells that eliminate infused cells. Such responses are associated with poor CAR T-cell expansion and relapse after initial remission. Evidence from MM trials using biepitopic CARs showed anti-CAR antibody development post-relapse, often coinciding with CAR T-cell clearance. Immune responses can arise from both humoral mechanisms and cross-presentation of murine peptides by antigen-presenting cells, which prime endogenous T cells to target the CAR [208].

6.4. Bispecific T-Cell Engagers (BiTEs)

BiTEs represent a new frontier in immunotherapy for RRMM, utilizing engineered antibodies to simultaneously bind a tumor antigen and CD3 on T cells, thereby redirecting cytotoxic T cells to eliminate malignant plasma cells (Figure 3). The most validated BiTE targets in MM are BCMA and the more recently identified G-protein-coupled receptor class C group 5 member D (GPRC5D). BCMA × CD3 BiTEs, such as teclistamab, were among the first to reach clinical application. In heavily pretreated RRMM patients, teclistamab demonstrated response rates of approximately 60% in early-phase trials, including high-risk subgroups [209]. These agents activate T cells independent of costimulatory signals and are highly potent even in patients previously treated with CAR-T or anti-CD38 antibodies [210]. Major adverse events include CRS and cytopenias, which are generally manageable with corticosteroids and IL-6 inhibitors. However, relapse after BCMA-targeted therapies is often driven by antigen escape or downregulation, prompting development of agents targeting GPRC5D, an antigen broadly expressed on MM cells but limited in normal tissues. Talquetamab, a GPRC5D × CD3 BiTE, has shown robust preclinical and clinical efficacy. Preclinical studies using constructs such as BR109 demonstrated potent, selective T-cell-mediated cytotoxicity in vitro and in xenograft mouse models, validating GPRC5D as a promising alternative to BCMA [211]. Moreover, GPRC5D-targeting BiTEs may overcome resistance seen in BCMA-exposed patients. Data indicate GPRC5D is more consistently and homogeneously expressed than BCMA, even in post-CAR-T or BiTE relapse settings [212]. Clinical development is ongoing for several GPRC5D × CD3 agents, including forimtamig and MBS314, which also show potential in targeting dual-negative clones [213,214]. To combat tumor heterogeneity and antigen escape, next-generation constructs are exploring trispecific formats, combining BCMA, GPRC5D, and CD3 arms. Agents like JNJ-79635322 and GBD218 have shown the ability to target dual-positive and single-positive cells, enhance T-cell engagement, and reduce cytokine release, all while retaining potent in vivo antitumor activity [215,216].

7. Future Directions in Therapeutics

7.1. Targeting Other Mutational Pathways

MAPK pathway, particularly the RAS/RAF/MEK/ERK axis, plays a central role in the regulation of cellular proliferation, differentiation, and survival. Mutations that activate this pathway are highly prevalent in MM, with NRAS, KRAS, and BRAF mutations collectively found in up to 50% of newly diagnosed MM cases. These alterations lead to constitutive signaling and are strongly associated with disease progression, drug resistance, and poor prognosis [217]. Given its role as a key oncogenic driver, the MAPK pathway has become a promising therapeutic target. MEK inhibitors, such as trametinib and selumetinib, have demonstrated cytotoxic effects against MM cells, especially those harboring activating RAS or BRAF mutations. MEK inhibition alone, however, often produces only modest effects due to compensatory signaling through parallel survival pathways [218]. A more effective strategy may involve combination therapy targeting both MAPK and PI3K/AKT pathways, which frequently co-activate in MM. In primary MM samples, dual inhibition of MEK/MAPK and PI3K/AKT signaling resulted in synergistic cytotoxicity in over 75% of cases, particularly those with RAS mutations [219]. In a biomarker-driven phase II clinical trial, trametinib was tested in RRMM patients stratified by RAS/RAF mutation status. Among 25 enrolled patients, the ORR to trametinib monotherapy was only 8%, with 17% achieving clinical benefit. However, when combined with the AKT inhibitor GSK2141795, the ORR improved to 27%, suggesting that co-targeting the MAPK and PI3K/AKT pathways may disrupt feedback loops and circumvent adaptive resistance mechanisms [220]. In the trial, thrombocytopenia and gastrointestinal symptoms were among the most common adverse events, but overall tolerability was acceptable. Notably, while the study did not demonstrate strong single-agent efficacy, it provided proof-of-concept that dual pathway blockade can enhance cytotoxicity in MM [220].

Targeting the p38 MAPK subpathway, which modulates the tumor microenvironment and cytokine signaling, is another emerging strategy. Inhibition of p38α MAPK using agents like VX-745 and SD-282 significantly reduced MM cell proliferation, IL-6 secretion, and tumor angiogenesis, particularly in the bone marrow niche [221,222]. More recently, dual inhibition strategies targeting MEK and MAP4K2—kinase-regulating stress-activated MAPK branches—have shown potent synergistic effects in RAS-mutated MM cells, including profound suppression of IKZF1/c-MYC and enhanced apoptosis [223]. This approach may provide a framework for personalized therapy in genomically defined subgroups. Furthermore, crosstalk between the MAPK pathway and DDR suggests that co-targeting these networks may enhance genomic stress and sensitize MM cells to apoptosis, especially in relapsed/refractory disease [224].

Next, synthetic lethality is a promising therapeutic approach that exploits the unique vulnerabilities of cancer cells harboring DNA repair defects. In MM, where genomic instability and defective homologous recombination (HR) repair are common, targeting compensatory DNA repair pathways—particularly with poly (ADP-ribose) polymerase (PARP) inhibitors—has gained momentum. MM cells often exhibit intrinsic chromosomal instability and dependence on HR to maintain survival, especially given frequent deficiencies in the non-homologous end-joining (NHEJ) pathway. This makes them uniquely susceptible to synthetic lethality strategies. PARP inhibitors, such as olaparib and veliparib, prevent repair of single-strand breaks, leading to accumulation of lethal double-strand breaks during DNA replication, particularly when HR is impaired [225]. Pharmacologic induction of “BRCAness” is another novel approach being explored. CDK inhibitors such as dinaciclib and THZ531 downregulate BRCA1/2 and RAD51 expression, mimicking HR deficiency and sensitizing MM cells to PARP inhibitors. In vitro and in vivo studies combining CDK inhibitors with PARP inhibitors (e.g., olaparib or ABT-888) have demonstrated synergistic cytotoxicity and tumor regression while sparing normal B cells [69,225]. In addition to PARP, ATR inhibition represents another synthetic lethality strategy. ATR is essential for responding to replication stress and DNA cross-links. In preclinical models, combining ATR inhibitors (e.g., VX-970) with melphalan, a frontline MM alkylator, led to synergistic anti-myeloma effects, even in melphalan-resistant lines. This strategy takes advantage of MM cells’ addiction to ATR signaling and has been shown to significantly prolong survival in xenograft models [226]. Moreover, combination therapies of PARP inhibitors with PIs like bortezomib have shown cell-line-specific synergy. In certain MM models, veliparib enhanced bortezomib cytotoxicity by reducing its effective dose sevenfold—suggesting that DNA damage responses modulate proteasome stress pathways [227]. These synthetic lethality strategies are especially attractive for HRMM or RRMM patients who harbor HR repair gene mutations (e.g., TP53, ATM, BRCA1/2) or show treatment resistance. Ongoing trials are needed to determine optimal biomarker-driven patient selection, dosing, and safety in combination with immunotherapy or cytotoxic agents.

7.2. Epigenetic Modulators

Epigenetic dysregulation is a hallmark of MM progression and therapeutic resistance. Abnormal expression of chromatin modifiers—such as BET proteins, HDACs, and EZH2—has been implicated in maintaining oncogenic transcriptional programs, particularly through upregulation of MYC, BCL-2, and other survival pathways. Thus, targeting these epigenetic regulators represents a promising approach for overcoming drug resistance and improving disease control in MM. BET inhibitors (e.g., JQ1, BAY 1238097) suppress transcription by preventing BET protein (such as BRD4) binding to acetylated histones. This leads to downregulation of oncogenes like MYC, BCL-xL, and cyclin D1. BET inhibition alone reduces MM cell proliferation and induces apoptosis, and it has shown synergistic effects when combined with other epigenetic agents [228]. However, resistance can arise through compensatory HDAC6 upregulation, suggesting the need for rational combination therapies [89]. HDAC inhibitors, such as panobinostat and vorinostat, block histone deacetylation, restoring transcriptional activation of tumor suppressor genes and triggering apoptosis. Panobinostat has been FDA-approved in combination with bortezomib and dexamethasone for RRMM. Preclinical data show that selective HDAC6 inhibition (e.g., ricolinostat) may enhance the efficacy of BET inhibitors like JQ1 by restoring MYC suppression and promoting apoptosis [229]. Importantly, dual HDAC/BET inhibitors are in development and show greater efficacy than either agent alone, with reduced off-target toxicity in preclinical models [230]. EZH2, a methyltransferase in the PRC2 complex, mediates trimethylation of histone H3K27 and represses gene expression. EZH2 is frequently overexpressed in MM and associated with poor prognosis and cell cycle dysregulation [86]. Although EZH2 inhibitors (e.g., GSK126, EPZ-6438) have limited single-agent activity in MM, their combination with HDAC inhibitors (like panobinostat) results in synergistic cytotoxicity and extensive transcriptional reprogramming that promotes apoptosis [87]. Furthermore, triple combinations targeting BMI-1 (PRC1), EZH2 (PRC2), and BET proteins have shown additive or synergistic anti-myeloma effects, enhancing chromatin accessibility and inducing durable apoptosis in vitro [90].

7.3. Vaccine Strategies and Tumor Neoantigens

Cancer vaccines targeting tumor-specific neoantigens represent an emerging and highly personalized immunotherapeutic strategy in MM (

Table 17. Investigational and available cancer vaccines in multiple myeloma.

Vaccine Name/Platform	Antigen Target(s)	Vaccine Type	Clinical Phase/Status	Mechanism of Action	Key Studies/Notes
PVX-410	XBP1, CD138, CS1 (SLAMF7)	Peptide-based multi-epitope vaccine	Phase I (completed)	Induces CD8+ T-cell responses against MM-associated antigens	Safe and immunogenic in SMM; studied alone and with lenalidomide (NCT01718899)
Bcl-2 family peptide vaccine	Bcl-2, Bcl-XL, Mcl-1	Peptide-based vaccine	Phase I	Stimulates cytotoxic T cells targeting anti-apoptotic proteins	Combined with bortezomib; immunogenic with minimal toxicity
MAGE-A3 peptide vaccine	MAGE-A3	Tumor antigen-specific peptide vaccine	Early phase/exploratory	Targets cancer–testis antigen expressed in MM and other malignancies	Used in pilot studies and as a possible post-ASCT strategy
Idiotype (Id) protein vaccine	Patient-specific paraprotein (Id)	Personalized protein vaccine	Phase I/II (historical interest)	Induces helper and cytotoxic T-cell responses against myeloma idiotype	Limited efficacy; used in post-ASCT settings; declining interest
DC/MM fusion vaccine	Whole-tumor antigens from patient myeloma cells	Dendritic cell (DC)-tumor fusion vaccine	Phase I/II	Enhances antigen presentation by fusing DCs with autologous MM cells	Shown to induce tumor-specific immunity post-ASCT (NCT02728102, NCT01067287)
Neoantigen-based vaccines	Patient-specific neoantigens	Personalized peptide/RNA vaccine	Preclinical/early phase	Stimulates T cells against mutation-specific targets unique to each patient	Requires whole-exome sequencing and epitope prediction
Gp96-Ig vaccine (ImPACT®)	Broad tumor antigen spectrum	Heat shock protein-based chaperone vaccine	Preclinical/exploratory	Cross-presents multiple antigens via HLA to CD8+ T cells	Used in MM and other hematologic malignancies
mRNA-based vaccines (exploratory)	Personalized neoantigens or MM-associated antigens	mRNA encoding tumor antigens	Preclinical	Induces cellular immunity via in vivo expression of encoded antigens	Emerging platform post-COVID-19 vaccine success
WT1 peptide vaccine	Wilms’ tumor 1 (WT1)	Peptide vaccine	Pilot/early phase	Targets overexpressed WT1 protein in MM	Studied in combination with other immunotherapies
NY-ESO-1 vaccine	NY-ESO-1	Cancer–testis antigen peptide vaccine	Early phase	Induces T-cell response against highly immunogenic antigen	Under study in hematologic and solid tumors

), particularly as a complement to existing T-cell-based therapies. Neoantigens—arising from somatic mutations in tumor cells—are not present in normal tissue and therefore evade central tolerance, making them ideal immunologic targets for vaccine development. Preclinical and early clinical studies in MM have confirmed that neoantigen-based vaccines can elicit both CD4+ and CD8+ T-cell responses, with protective and therapeutic potential. In a mouse model of MM (MOPC315), a vaccine composed of an idiotype (Id) neoantigen fused with a heat shock binding motif and poly (I:C) adjuvant induced strong CD4+ T-helper responses, which in turn activated CD8+ cytotoxic T cells against non-vaccine tumor-associated antigens, resulting in tumor control and immune memory [231]. Neoantigen vaccines are now entering clinical trials. In the PGV-001 phase I trial, a multipeptide personalized neoantigen vaccine was administered to patients (including those with MM) in the post-remission setting. Each patient’s tumor underwent genomic sequencing to identify 8–10 predicted MHC class I-binding neoantigens. The vaccine induced neoantigen-specific CD4 and CD8 T-cell responses, was well tolerated, and confirmed the feasibility of individualized peptide vaccine synthesis in MM [232,233]. Furthermore, neoantigen-specific T-cell responses have now been validated directly in MM patients. A study analyzing tumor sequencing data from 184 MM patients identified shared immunogenic mutations in driver genes like KRAS, NRAS, and IRF4 and demonstrated that neoantigen-specific CD8+ T cells correlated with tumor regression in relapsed patients [234]. These results suggest that even in cancers with intermediate mutational burden, like MM, neoantigen-based vaccines are feasible and may enhance immune surveillance.

Next, a phase I clinical trial evaluated the safety and immunogenicity of therapeutic peptide vaccines targeting Bcl-2 family proteins (Bcl-2, Bcl-XL, Mcl-1) in patients with relapsed MM. Seven patients received vaccinations alongside bortezomib, with four completing the full eight-dose schedule. The vaccine was well tolerated, with no unexpected toxicities beyond those associated with bortezomib. Among the six patients who received more than two doses, immune responses to the peptides were observed in all, and three showed increased immune reactivity after vaccination. These results suggest that Bcl-2-targeted peptide vaccination is feasible, safe, and immunologically active in relapsed MM, supporting further investigation in larger trials [235]. Another study evaluated the immunologic potential of a multipeptide vaccine cocktail composed of four HLA-A2–restricted peptides derived from MM-associated antigens—XBP1 (US184–192 and SP367–375), CD138 (260–268), and CS1 (239–247). Cytotoxic T lymphocytes (CTLs) generated from HLA-A2+ donors using this peptide cocktail demonstrated robust, antigen-specific immune responses, including increased populations of effector memory (CCR7⁻CD45RO⁺) and activated (CD69⁺) CD8⁺ T cells. The multipronged CTLs also showed proliferation, IFN-γ production, and cytotoxic activity against HLA-A2+ MM cell lines and primary MM patient cells, while sparing irrelevant targets. Importantly, each peptide within the cocktail contributed to antigen-specific activity without compromising overall immune function. These findings support the feasibility of using multiantigen peptide vaccines to generate broad and potent MM-specific T-cell responses [236].

Innovative formulations are under investigation to improve immune activation. For example, synthetic peptide vaccines incorporating heat shock motifs or delivered via viral or nanoparticle platforms are being tested to improve antigen presentation and elicit broader, durable T-cell responses [237]. Combinations with checkpoint inhibitors or IMiDs are also being explored to overcome T-cell exhaustion and enhance vaccine potency [238]. One challenge remains the heterogeneity of neoantigens between patients and within tumor subclones, which necessitates highly personalized vaccine design. Advances in sequencing and neoantigen prediction pipelines (e.g., OpenVax) are enabling rapid identification and ranking of immunogenic targets for vaccine synthesis.

7.4. Microenvironment Modulation

The bone marrow microenvironment in MM plays a pivotal role in promoting tumor progression, immune evasion, and therapy resistance. Key players in this immunosuppressive niche include myeloid-derived suppressor cells (MDSCs), Tregs, and dysregulated cytokine networks. These elements foster a tumor-promoting immune landscape, thereby undermining the efficacy of immunotherapies and contributing to relapse. MDSCs are significantly expanded in both the peripheral blood and bone marrow of MM patients, particularly in those with active disease. They not only enhance MM cell survival and proliferation but also suppress T-cell responses via arginase-1, reactive oxygen species (ROS), and inhibitory cytokines like IL-6 and IL-10 [239]. Furthermore, MDSCs facilitate the induction of Tregs and impair dendritic cell function, leading to broader immunosuppression and reduced antitumor surveillance [240]. Therapeutic strategies targeting MDSCs are under active investigation. Notably, the multi-kinase inhibitor sunitinib has shown efficacy in suppressing both monocytic and neutrophilic MDSC subsets, restoring T-cell proliferation and enhancing the cytotoxic effects of lenalidomide in co-culture models [241]. Additionally, HDAC inhibitors and phosphodiesterase-5 (PDE5) inhibitors are being evaluated for their capacity to deplete or reprogram MDSCs [242]. Tregs also accumulate in MM and correlate with poor prognosis. These FoxP3+ cells suppress effector T-cell activation and facilitate immune escape. Targeting Tregs—either through IL-2-diphtheria toxin fusion proteins (e.g., E7777) or checkpoint blockade combinations—has shown promise in reducing immune suppression and enhancing tumor-specific responses [243]. Moreover, cytokine dysregulation, especially overproduction of IL-6, IL-10, TGF-β, and prostaglandin E2 (PGE2), reinforces the immunosuppressive loop within the bone marrow microenvironment. IL-6 promotes plasma cell survival and drug resistance, while IL-10 and TGF-β inhibit cytotoxic T cells and dendritic cell function. Targeting these cytokines or their receptors (e.g., anti-IL-6 monoclonal antibodies or EP4 antagonists) is being explored to restore immune balance and improve responses to immunotherapy [243,244].

7.5. Precision Medicine and Biomarker-Driven Trials

Precision medicine in MM aims to tailor therapies based on individual molecular and clinical profiles, moving beyond the traditional “one-size-fits-all” paradigm. With the rise of NGS and multi-omics profiling, clinical trial designs have evolved to accommodate this complexity. Among these, basket and umbrella trials (Figure 4)—core master protocol strategies—enable more adaptive and biomarker-driven evaluation of novel therapies. Basket trials enroll patients with different tumor types but a common molecular alteration, allowing researchers to assess whether a therapy targeting that alteration is effective regardless of the cancer’s tissue origin. In contrast, umbrella trials focus on a single tumor type (e.g., MM) but test multiple therapies matched to different molecular subtypes within that disease. These trial designs allow efficient stratification, faster drug development, and more personalized therapy allocation [245,246]. In MM, where inter- and intra-patient heterogeneity is high, umbrella trials could enable simultaneous testing of targeted agents against genomic aberrations like t(11;14), del(17p), or RAS mutations in defined subgroups. However, the success of these designs hinges on a clear understanding of biomarker classification. Prognostic biomarkers provide information about a patient’s likely disease outcome independent of treatment. For example, del(17p) and gain(1q21) are negative prognostic markers in MM, predicting early relapse and shorter OS. In contrast, predictive biomarkers indicate response likelihood to a specific therapy, enabling patient selection for targeted treatments. BCL2 expression in t(11;14) myeloma, which predicts response to venetoclax, is a classic example of a predictive biomarker [247].

Many biomarker-driven trials now incorporate adaptive enrichment strategies, in which ongoing biomarker assessments refine trial arms and eligibility as data accumulates. These dynamic designs are critical in MM, where clonal evolution may alter mutational profiles over time. Additionally, platform trials—a hybrid of basket and umbrella designs—allow continuous addition or closure of arms based on interim efficacy data, making them well-suited for rapidly evolving therapeutic landscapes like MM [248]. Despite these advances, challenges persist. The combinatorial complexity of MM mutations, subclonal heterogeneity, and overlap of predictive and prognostic functions complicate trial interpretation and therapy matching. Moreover, robust biomarker validation is often lacking in early-phase MM studies, limiting their generalizability [249].

8. Special Considerations

8.1. Management in Elderly and Frail Patients

MM is predominantly a disease of older adults, with a median age at diagnosis of approximately 70 years. As the aging population grows, so does the clinical challenge of managing MM in elderly and frail patients, who often present with multimorbidity, reduced functional reserve, cognitive decline, and polypharmacy. These factors contribute to increased treatment-related toxicity, early therapy discontinuation, and worse survival outcomes compared to younger or fitter patients [250,251]. Historically, treatment selection in elderly patients was based on chronological age and performance status. However, age alone is insufficient to predict tolerability. Instead, frailty assessments using tools such as IMWG frailty index and the Revised Myeloma Comorbidity Index (R-MCI) are now recommended. These tools stratify patients into fit, intermediate-fit, or frail, guiding treatment intensity accordingly [252]. Frailty-adapted therapy aims to balance efficacy with toxicity, prioritizing symptom control, quality of life, and functional independence. Fit elderly patients may receive full-dose triplet regimens like VRd, while frail individuals often benefit more from two-drug combinations or attenuated regimens, such as dose-reduced lenalidomide and dexamethasone (Rd-lite) [253]. Newer oral triplet regimens like ixazomib-Rd have also shown promise in frail patients, offering lower toxicity and improved quality of life [254].

Evidence from real-world studies and prospective trials supports the prognostic value of frailty assessments, not only at baseline but also as a dynamic metric during therapy. A large meta-analysis demonstrated that frailty independently predicts mortality and adverse events in MM regardless of treatment choice [255]. Moreover, routine geriatric reassessments can help adjust therapy over time and identify patients at risk of treatment discontinuation due to evolving impairments [256]. To facilitate implementation in daily practice, electronic and rapid-screening tools are being integrated into clinical workflows. These digital geriatric assessments require minimal time and have been shown to influence treatment decisions in over 30% of cases, with clinicians escalating or de-escalating therapy based on frailty status [257]. As such, comprehensive geriatric assessment (CGA) is essential in guiding individualized, frailty-adapted therapy for elderly MM patients. It improves therapeutic decision-making, minimizes toxicity, and aligns treatment with patient-centered goals, marking a critical step toward precision oncology in aging populations.

8.2. Racial and Ethnic Disparities

MM presents significant racial and ethnic disparities in both incidence and outcomes. African Americans (AAs) have a two- to threefold higher incidence of MM compared to White Americans, and they tend to be diagnosed at younger ages [258]. Despite this increased risk, some studies have found better or equivalent OS among AAs when provided with equal access to care, suggesting that non-biological factors, such as access to treatment and healthcare delivery systems, may be key contributors to outcome disparities [259]. Recent genomic studies have uncovered differences in disease biology that may partially explain disparities. AAs exhibit a higher frequency of the t(11;14) translocation and a lower prevalence of high-risk cytogenetic abnormalities such as del(17p), which could contribute to relatively favorable biology in this group [260]. Furthermore, data from the National Institute of Health (NIH) “All of Us” cohort demonstrated a synergistic interaction between African genetic ancestry and hypertension as a strong risk factor for MM, highlighting the role of gene–environment interplay in racial disparities [261].

Yet, the social determinants of health (SDOH) remain the dominant modifiable contributors to disparities. Socioeconomic status, insurance coverage, education level, and geographic access to specialized care profoundly affect outcomes. For example, Black and Hispanic patients are less likely to receive ASCT and novel agents like lenalidomide or monoclonal antibodies, even when clinically indicated [262,263]. These treatment gaps are further exacerbated by underrepresentation of minorities in clinical trials, limiting generalizability and access to cutting-edge therapies [264]. Moreover, Hispanic and Native American patients face additional barriers and often present with more advanced disease. A study using SEER data revealed that Hispanic patients had the worst OS, potentially due to both healthcare access issues and differences in disease biology [265]. Among Hispanics, younger age at diagnosis and higher rates of extramedullary disease were reported, but cytogenetic profiles did not clearly explain survival gaps, reinforcing the need to further explore healthcare inequities [266]. In conclusion, racial and ethnic disparities in MM reflect a complex interplay between genetic predisposition and socioeconomic disadvantage. Addressing these disparities will require not only a deeper understanding of disease biology across ancestral backgrounds but also systemic reform to ensure equitable access to diagnostics, therapies, and clinical trials.

8.3. Relapsed and Refractory MM

RRMM presents a significant therapeutic challenge, as nearly all patients eventually relapse despite initial responses to frontline therapy. The goal in RRMM is to maximize disease control while preserving quality of life through the rational use of sequential, non-cross-resistant therapies and incorporation of emerging agents. Therapy selection is increasingly individualized based on prior treatments, depth and duration of previous response, patient comorbidities, and relapse aggressiveness. For patients with early relapse (e.g., <12 months after autologous stem cell transplant or triple-class exposure), intensive triplet or quadruplet rescue regimens are preferred, often including next-generation PIs (carfilzomib), immunomodulatory drugs (pomalidomide), or monoclonal antibodies (daratumumab, isatuximab) [267]. Patients with indolent relapse or frailty may benefit from reusing previously effective regimens, particularly if the treatment-free interval exceeded 18 months [268]. Sequential therapy, rather than upfront combination of all active agents, remains a viable strategy in certain scenarios, particularly for chronic management of RRMM. This approach reserves newer agents for future lines, potentially reducing cumulative toxicity and extending overall treatment durability. However, combination therapy in early relapse may lead to deeper responses and prolonged survival—highlighting the need to balance treatment intensity with patient-specific factors [269]. Clinical trials are pivotal in the RRMM setting, especially for patients who are refractory to PIs, IMiDs, and anti-CD38 antibodies—so-called triple-class refractory disease. Participation in trials offers access to novel mechanisms of action, such as bispecific antibodies, CAR-T cells, ADCs (e.g., belantamab mafodotin), and CELMoDs, many of which have demonstrated promising activity in early-phase studies [270]. For example, patients enrolled in early trials combining lenalidomide and bortezomib achieved durable responses with manageable toxicity even after multiple relapses [271]. Importantly, clinical trial participation improves survival outcomes compared to standard therapies alone and should be strongly encouraged, especially when optimal treatment sequences remain undefined [272]. With an increasing number of trials incorporating biomarker stratification and real-time molecular monitoring, future RRMM therapy will likely become even more personalized and adaptive.

9. Conclusions

MM represents a biologically complex and clinically heterogeneous malignancy characterized by progressive genetic and epigenetic alterations, immune escape, and microenvironmental crosstalk. Over the past two decades, advances in molecular diagnostics, risk stratification, and the development of targeted and immune-based therapies have significantly improved patient outcomes. However, MM remains incurable for most patients, with inevitable relapse and the emergence of drug resistance posing ongoing challenges. Looking ahead, future strategies must focus on integrating genomic and immune profiling into real-time clinical decision-making to enable personalized therapy selection. Emphasis should be placed on developing biomarker-driven clinical trials, especially adaptive basket and umbrella designs, to match emerging agents with molecularly defined subgroups. The microenvironment and mechanisms of immune suppression, including the roles of MDSCs, Tregs, and cytokine networks, remain promising therapeutic targets. Moreover, optimizing the use of CAR-T cells, bispecific T-cell engagers, and vaccine-based approaches in combination regimens or earlier disease stages holds potential for deeper and more durable responses. As treatment becomes increasingly complex, frailty assessments, equity in access, and inclusive trial enrollment will be critical to ensure that novel advances benefit all patient populations. Ultimately, a multidisciplinary, precision-guided approach is essential to transform MM into a manageable—and eventually curable—disease.