SUMMARY

• Multiple myeloma (MM) is a malignant cancer where an increased number of clonal plasma cells are found in the bone marrow (BM). Even after high-dose therapy followed by stem cell transplantation, recurrence is still the leading cause of death. Minimal residual disease (MRD) is the persistence of residual tumor cells which are responsible for relapse.¹
• Changes in the level of serum paraprotein and/or urinary light-chain excretion form the basis of assessing response to therapy and monitoring progress of MM.²
• In newly-diagnosed high-risk, and relapse/refractory multiple myeloma (RRMM) patients, there is a direct correlation between the depth of response, including complete response (CR), and progression-free survival (PFS) and overall survival (OS).²
• The presence of MRD, even among CR patients, is frequently an adverse prognostic feature, with high MRD levels correlating with a shorter time to progression, and a deeper level of MRD-negativity being linked to better survival.^2,3
• Techniques to monitor intramedullary MRD include multiparameter flow cytometry (MFC), allele-specific oligonucleotide, polymerase chain reaction (ASO-PCR), nextgeneration sequencing (NGS), and nextgeneration flow (NGF). Positron emission tomography-computed tomography (PET/CT) is used to detect extramedullary MRD.²
• Multiple studies have shown the value of using MRD as a biomarker to evaluate the efficacy of treatment phases, to support potential treatment decisions, and to act as a surrogate for OS.^2-4
• Additional studies highlighting the importance of MRD in MM have also been cited.^5-10

BACKGROUND

Historical Overview

For some potentially curable neoplasms, the determination of MRD is becoming standard diagnostic care. In spite of achieving progressively higher CR rates, patients with MM will inevitably relapse.³ Assessing response to therapy and monitoring the progress of MM was based on the changes in the level of serum paraprotein and/or urinary light-chain excretion.²

The prognostic value of CR has been validated in both transplant-candidate and elderly patients, and a correlation between deeper quality of responses and better outcomes has been described in the relapsed/refractory (R/R) setting. The maintenance of CR has been shown to be more important than the achievement of CR, as patients who relapse from CR early-on have been shown to have unfavorable outcomes. Taken together, CR can be a strong prognostic marker and a clinically relevant endpoint.²

In 2006, the International Myeloma Working Group (IMWG) introduced normalization of serum free light-chains (sFLC) and absence of clonal plasma cells in BM biopsies by immunohistochemistry and/or immunofluorescence as additional requirements to define stringent CR.¹¹ One large study was able to show superiority of the stringent CR over conventional CR criteria to define patients’ outcomes, while other groups failed to demonstrate the utility of the sFLC assay among immunofixation-negative patients. Additionally, the majority of patients who achieve CR after therapy show recovery of normal plasma cells that exceeds the percentage of clonal plasma cells, implying that more sensitive clonality markers are needed (eg, clonotypic immunoglobulin [Ig] gene sequences or immunophenotyping). The incorporation of immunophenotypic and molecular CR as part of the response criteria in MM must be matched by highly sensitive technologies able to detect MRD at very low levels.²

Rationale for MRD in MM

As demonstrated in individual studies and meta-analyses among transplant-eligible and transplant-ineligible patients, newly-diagnosed high-risk, and RRMM patients, there is a direct correlation between the depth of response, including CR, and PFS and OS.²

Although in many cases, patients who achieve deeper response have prolonged survival, there have been instances in which response rates were not a direct surrogate of survival outcomes, for example: patients in CR relapsing early with dismal survival and patients who failed to achieve CR showing favorable outcomes. Incongruous results may be related to: heterogeneity of consolidation or maintenance treatment used in a treatment arm after response evaluation which may affect tumor reduction, or different CR quality reached after different regimens, combined with limited sensitivity of current methods to define CR. These observations highlight the need for standardization and optimization of MRD detection.²

The presence of MRD is frequently an adverse prognostic feature, with high MRD levels correlating with a shorter time to progression, and a deeper level of MRD eradication being linked to better survival.²

IMWG MRD Criteria

The current IMWG response criteria for MRD details variations of MRD present in the BM, building on existing IMWG response criteria (see Table: IMWG MRD Criteria).¹²

When MRD results are reported, the assessment should be qualified by the method(s) used (flow MRD-negative or sequencing MRD-negative) and the level of sensitivity (eg, 1 in 10⁵ or 1 in 10⁶ cells).¹²

Methods Used to Monitor MRD

Sequencing-Based Methods

NGS

NGS can quickly perform multiple reads of many different DNA fragments, detecting previously known tumor-specific sequences within normal DNA fragments. The sensitivity of NGS is estimated to be in the range of 10^-5-10^-6. NGS is methodically less complex than ASO-PCR and can identify the variability of normal polyclonal B-cells.² Phase 3 MM trials data wherein MRD has been evaluated by NGS is presented in Table: Phase 3 Clinical Trials Incorporating MRD Evaluation by NGS in MM.¹³

Flow Cytometry-Based Methods

NGF

NGF combines the analysis of 10 markers (cluster of differentiation [CD]38, CD138, CD45, CD19, CD27, CD56, CD81, CD117, cytoplasmic Igκ, and cytoplasmic Igλ) in 2 independent tubes analyzed by 8-color MFC. The markers included in the panel allow distinction between normal and clonal neoplastic plasma cell. NGF is also useful for the evaluation of other cell compartments in the BM. The EuroFlow’s NGF panel reaches a maximum sensitivity of 2×10^-6 when at least 10 million events are analyzed, and it is currently recommended by the IMWG as a reference immunophenotyping method for the evaluation of MRD in MM.¹³

The IMWG-defined response criteria for MRD-negativity cutoff is 10^-5, detected either by NGS or NGF. A general trend in increasing the MRD sensitivity to 10^-6 has been observed in clinical trial designs, which can be a better predictor of PFS.¹⁴

MRD evaluation via NGF and NGS should be done at treatment cessation, 3 months after autologous stem cell transplantation, and every 6 months thereafter, at least for 2 years if negativity is achieved. This systematic evaluation could reveal a sustained MRD-negativity status that is crucial for long-term remission.¹⁴

A high degree of difference is observed in NGF in terms of BM cellularity and percentage of plasma cells by cytological analysis compared with flow cytometry methods owing to the fragility of plasma cells themselves, with the loss of plasma cells during laboratory manipulation. Additionally, the lower plasma cell count obtained by flow cytometry may be related to a possible hemodilution of the BM samples, with the risk to underestimate the percentage of pathologic plasma cells.¹⁴

The high heterogeneity of MM plasma cells and the potential for a “shift” in plasma cells phenotype depending on the therapy that patients have received are both major disadvantages in NGF’s evaluation of MM MRD. Drugs such as DARZALEX that mask the CD38 overexpressed molecule can also make MM clones’ recognition more difficult. This problem is overcome by introducing CD38 multiepitope antibodies that allow recognition of MM plasma cells even during treatment with anti-CD38 monoclonal antibodies by binding to sites that are different from the one occupied by the drug.¹⁴

Given the peculiarity of flow cytometry analysis, a high personnel expertise is essential to obtain reliable results, especially in demanding cases in which anti-CD38 therapy, presence of different pathologic clones, or presence of normal plasma cells, together with low MRD burden, could lead to bias.¹⁴

MFC

MFC allows simultaneous identification and characterization of normal vs tumor cells at the single-cell level, evaluation of a high-cell numbers over a few hours, quantitative assessment of normal and tumor cells and their corresponding antigen expression levels, combined detection of cell surface and intracellular antigens, and an overview of hematopoiesis through analysis of different cell lineages.²

Molecular-Based Methods

ASO-PCR

ASO-PCR bypasses immediate sample processing, as it is unaffected by loss of viable cells over time, making it a possibility for studies requiring centralized or delayed analysis. Polymerase chain reaction (PCR) assays can reach highly sensitive MRD detection levels up to 10^-6, with routine limits of detection at 10^-5. PCR methods have been standardized, validated, and reproduced for MRD testing in other hematological neoplasms (eg, acute lymphoblastic leukemia), though not yet in MM.²

Droplet Digital Polymerase Chain Reaction

Droplet digital polymerase chain reaction (ddPCR) increases the chance of identifying molecular targets (such as Ig heavy chain genes IGH rearrangements, fusion genes, or mutations) to be followed after therapy, thus making it a promising tool in MRD assessment. It is based on target compartmentalization, endpoint PCR, and Poisson statistics. Compared with traditional quantitative polymerase chain reaction (qPCR) approaches, ddPCR is more precise, better at detecting rare genetic events, and less susceptible to inhibitors. In addition, ddPCR does not require a standard curve, presenting a practical advantage over qPCR. Thus, this method does not have any fluctuations in reaction efficiency and makes MRD detection possible for all those patients in whom generating a standard curve is not feasible.¹³

Imaging-Based Methods

PET/CT

PET/CT, which combines the morphological images provided by computed tomography (CT) with imaging data of a metabolic process (eg, fluorodeoxyglucose uptake) is the technique of choice to detect extramedullary disease. Both false positive (eg, coexisting infectious or inflammatory processes) and false negative results (eg, quiescent tumor cells) may occur. PET/CT is the most promising imaging tool to monitor MRD in MM. Standardization of PET/CT and comparison with other sensitive BM-based MRD methods is needed to implement imaging monitoring in the clinical setting.²

Blood-Based Methods

M-Protein Detection Using Intact Immunoglobulin Methods (Intact Protein mass spectrometry [MS])

Intact protein MS measures light-chain mass/charge ratio (m/z) using matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF), which can reach high throughput and is easier to standardize. This method requires a higher concentration of M-protein. This assay is suitable for routine diagnostics because of its rapid turnaround time and its high specificity. Using a 15-minute gradient, the intact protein MS method can reach a concentration of 250 mg/L for detecting the M-protein heavy chain. By increasing the separation time, this method, termed monoclonal Ig rapid accurate molecular mass, can improve the lower limit of detection (LLOD) by almost 10 times, to 5 mg/L.¹⁵

M-Protein Quantification Using Clonotypic Peptide Methods (Bottom-Up MS)

Personalized targeting of monoclonal peptides using bottom-up MS is technically complex but increases the sensitivity of the assay. Using a 15-minute gradient, the bottom-up method can reach an LLOD of 43 mg/L.¹⁵

Techniques to monitor intramedullary and extramedullary MRD are compared in Table: Individual Features of Currently Available Techniques to Monitor MRD in Multiple Myeloma.

While both techniques provide similar prognostic information, ASO-PCR is slightly more sensitive and specific than MFC for the evaluation of MRD. However, ASO-PCR is applicable in a lower proportion of MM patients and is more costly and time-consuming.¹ NGS techniques are a valid alternative to ASO-PCR methods with reduced complexity and rates. NGS can rapidly identify the most current mutations in MM, translocation breakpoints, immunoglobulin heavy chain (IgH) isotype, IGH translocations, clonal rearrangements and specific sequences developed during isotype switching. This leads to the classification of molecular subgroups of patients and the data acquired may assist in the development of target therapy.¹⁶

Comparison between molecular- and flow-based approaches for MRD assessment in MM is presented in Table: Current Molecular and Immunophenotyping MRD Approaches.

Incorporating MRD Into Clinical Studies

Studies have shown the value of using MRD as a biomarker to evaluate the efficacy of treatment phases, to support potential treatment decisions (eg, chemosensitive vs chemoresistance before/after high-dose therapy/autologous stem cell transplantation), and to act as a surrogate for OS.^2-4,16 Research is needed to determine how to standardize techniques and integrate medullary and extramedullary MRD monitoring.^2,3,16

Optimal Timepoint for MRD Evaluation

In clinical trials evaluating previously untreated patients, MRD is usually monitored at the end of induction, following transplantation in healthy subjects (3 months is the most common timepoint), and periodically thereafter.¹³

A single MRD-negative result in a patient can also be achieved with several drug combinations. While this format of assessment still has clinical implications, MRD disappearance may be transient, so only the most beneficial treatment scheme (and lethal for MM cells) will result in a sustained clearance of tumor cells. Therefore, reporting sequential MRD-negativity rates is preferred in clinical trials as the best indicator of therapy effectiveness.¹³

MRD can also be performed after induction as well as 3- and 24-months after transplantation and can be used to describe how patients with sustained MRD-negativity for 2 years after the end of induction have similar outcomes to those becoming MRD-negative within 24 months; however, longer OS in patients with sustained MRD-positivity or MRD resurgence show similarity in outcomes to those reported in a phase 2 trial from the Memorial Sloan Kettering Cancer Center.¹³

In a pooled analysis of the MAIA and ALCYONE studies, MRD was assessed at ≥6 or ≥12 months. In the TOURMALINE-MM3 and -MM4 trials, MRD was assessed continuously as a prognostic indicator during the maintenance phase (MRD resurgence rate 9.5%, MRD-negativity rate 5.1%).¹³

MRD as a Prognostic Factor

A meta-analysis of 67 clinical studies revealed that achievement of MRD-negativity led to improved PFS and OS regardless of multiple patient- and disease-related factors (MRD sensitivity threshold, cytogenetic risk, method of MRD assessment, and depth of clinical response).⁴

A pooled analysis of 4 phase 3 clinical studies of DARZALEX in patients with MM (ALCYONE, CASTOR, MAIA, and POLLUX) revealed that patients who achieved ≥CR with MRD-negativity had improved PFS compared with those who did not achieve both.⁴

MRD as an Endpoint in Clinical Studies

Due to its strong correlation with prognostic factors (eg, PFS and OS), MRD-negativity has been popularly accepted as a surrogate endpoint in clinical studies.^2-4,16 To facilitate accelerated drug development, MRD has also been used in studies evaluating new combination therapies as an indicator of efficacy.⁴

MRD-adapted Treatment Strategy

Due to its correlation with long-term outcomes, MRD has been used to adapt/modify treatment strategies in several phase 3 (MIDAS, DRAMMATIC, AURIGA), phase 2 (MASTER, NCT04140162, DART4MM; NCT05192122, FREEDMM), and phase 2/3 (REMNANT) studies.^4,17,18

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File (and/or other resources, including internal/external databases) was conducted on 08 August 2024.