Усиливающиеся тенденции измерения минимальных остаточных заболеваний при испытаниях, сосредоточенных на многократном лечении миеломы: систематический анализ клинических исследований 2014 в адрес 2025.

Introduction

Minimal residual disease (MRD) has emerged as a central biomarker in multiple myeloma (MM), not only for its prognostic value but also for its growing role as a regulatory endpoint [ 1 ]. The accurate detection of MRD, via next‐generation flow cytometry (NGF) or next‐generation sequencing (NGS) at sensitivities reaching 10 −5 to 10 −6 , enables the identification of deep remission states associated with significantly improved progression‐free survival (PFS) and overall survival (OS) [ 2 , 3 ].

In parallel with the growing emphasis on MRD, conventional therapy response outcomes such as overall response rate (ORR), complete response (CR), and very good partial response (VGPR) remain central to evaluating treatment efficacy in MM trials. These endpoints, defined by International Myeloma Working Group (IMWG) criteria, are widely used to benchmark the activity of novel agents and combination regimens in both newly diagnosed and relapsed/refractory settings [ 4 ]. While high response rates are often seen as early indicators of therapeutic success, their correlation with long‐term outcomes can be variable.

In 2016, MRD was formally integrated into the IMWG response criteria, a milestone influenced by key opinion leaders in the field [ 4 ]. The IMWG, along with other international organizations, played a pivotal role in advocating MRD integration into clinical decision‐making and regulatory frameworks, noting that MRD status can guide therapeutic intensification or de‐escalation and may serve as an early surrogate for survival endpoints in trials [ 5 ].

Contemporary trial designs increasingly feature MRD endpoints, and many Phase 3 trials demonstrate higher MRD‐negative rates in experimental arms well before PFS differences emerge [ 6 ]. In addition, MRD is being adopted not only as exploratory or secondary endpoints but also as stratification factors or decision‐based triggers in adaptive treatment protocols [ 7 ]. In clinical practice, experts have begun incorporating MRD testing to help tailor treatment plans at various points of the disease course, although consensus on its routine use is still developing [ 8 ]. Imaging modalities such as PET/CT are increasingly combined with molecular MRD to better detect residual disease—especially extramedullary or patchy marrow involvement—enhancing risk stratification and potentially improving trial eligibility criteria and endpoints [ 9 , 10 ].

A recent publication in Nature Reviews Clinical Oncology provided detailed information on both the methodological strengths and ongoing challenges of MRD implementation in MM management, underscoring the need for assay standardization and consensus definitions across trials [ 11 ]. Their review also highlights emerging opportunities such as the consideration of MRD by regulatory agencies such as EMA and FDA as an early endpoint for accelerated approvals, a position that was presented already in various platforms [ 12 ].

At EMA, a qualification advice was requested in November 2024, on the use of MRD as an early endpoint in clinical trials in order to support regulatory decisions (see detailed in Supporting Information : Appendix 1 ). In their response, the Committee for Medicinal Products for Human Use (CHMP) agreed that MRD negativity at a 10 −5 or higher threshold has a prognostic value at patient level, and the available data appear to support this positive effect in different MM settings (i.e., newly diagnosed transplant‐eligible, newly diagnosed transplant‐ineligible, relapsed/refractory). CHMP also agreed that, depending on the setting, the MRD negativity endpoint can support (conditional) approval of a compound; however, the obligation to demonstrate long‐term benefit remains. In addition, they noted that the generalizability of MRD benefits across all new treatment modalities requires further investigation and emphasized that a high level of standardization in MRD testing has not been achieved yet. Therefore, guidance is needed to ensure that MRD assessments are consistent amongst laboratories. One of their key conclusions was that the suitability of MRD negativity at a pre‐specified timepoint as an endpoint in clinical trials in order to support regulatory decision making depends on the treatment setting and study design and will be based on a case‐by‐case assessment.

Despite these advances, relatively few studies have systematically characterized the adoption of MRD in MM trials over extended periods. A review of trials between 2015 and 2020 on ClinicalTrials.gov revealed wide heterogeneity in MRD methodology, low rates of trials using MRD as a primary endpoint or to directly guide therapy, and minimal inclusion of imaging‐based MRD [ 13 ].

To contribute to the existing knowledge in this field, the purpose of this study was to investigate the MM trials that have been initiated over the past more than 10 years, with special focus on the MRD and response outcomes. The analysis also aims to benchmark the trial design evolution against emerging regulatory expectations and the EMA/CHMP position used as a reference framework to interpret the maturity, strengths, and gaps of the current MRD evidence ecosystem.

Methodology

In this research, the database of clinicaltrials.gov was searched on July 11th, 2025, from January 1st, 2014, to cover the past over 10 years period. The condition of “MM” was searched without any limitation for interventions, locations or study status. The full dataset was exported into Microsoft Excel for the further analysis. Both interventional and observational studies were included in the original dataset.

The downloaded database was screened for interventional trials or observational studies specifically focusing on therapeutic alternatives for MM patients. During the screening process, the following exclusion criteria were applied: (1) studies investigated other disease areas in hematology; (2) studies focused on the therapy of a specific condition associated with MM (e.g., an adverse event); (3) studies with non‐curative therapeutic alternatives. Studies passing these exclusion criteria were considered relevant for inclusion and data extraction.

The data extraction was also conducted in Microsoft Excel. First, we reviewed whether the therapy response or the MRD were listed as outcomes in the studies based on the outcome measure and the outcome description parameters. Then we identified if these outcomes were primary or secondary. Further information was collected on the details of MRD measurement: the technology applied for the assessment (i.e., NGS, flow cytometry or other types); the cut‐off for MRD‐negativity (e.g., 10 −4 or 10 −5 ); the timing of MRD assessment (e.g., during or after therapy, during or after a specific follow‐up period). In addition, we also identified those studies, where the results of the MRD assessment was used to guide the further therapeutic regimens.

The screening process and data collection on the downloaded dataset were conducted and confirmed by two researchers independently. In this paper we mainly focus on the MRD outcome parameters and present the results annually to investigate the key trends of the trials from the past over 10 years period. Importantly, for 2025, we only present approximately for half a year.

Results

A total of 2064 studies were included in the dataset. After the screening, 728 were considered not relevant for further analysis, as they did not investigate a specific MM therapy. From the 1336 studies, 1209 were interventional clinical trials and 127 were observational studies. In the following parts of the results section, first we present findings from the clinical trials, then separately from the observational studies.

From the included clinical trials, 22% ( n = 266) had the status of completed. Five trials were suspended, 12% ( n = 142) were terminated, 6% ( n = 70) were withdrawn and 11% ( n = 135) had unknown status. A total of 25% ( n = 308) were in the recruiting status, 1 had the status of enrolling by invitation, 17% ( n = 205) were active, but not recruiting, while 6% ( n = 77) were not yet recruiting. The number of trials according to the year of initiation ranged between 74 (in 2025, where only half a year was considered) and 129 (in 2021). The phase of the trials was as follows: Phase 4 ( n = 25); Phase 3 ( n = 127); Phase 2 and 3 ( n = 9), Phase 2 ( n = 429); Phase 1 and 2 ( n = 197); Phase 1 ( n = 326); early Phase 1 ( n = 53); not available ( n = 43). Overview of the trial characteristics by year is shown in Tables 1 and 2 .

In the total sample, response was included in 86.4% ( n = 1044) of the trials, while MRD was included in 30.9% of the trials ( n = 373). The trend of these outcomes is shown on Figure 1 . Incorporation of response remained fairly stable over time with the lowest in 2014 (77.8%) and highest in 2025 (94.6%). On the other hand, MRD increased steadily over time with the lowest in 2014 (6.7%) and the highest in 2025 (56.8%).

Proportion of outcomes by trial status is presented on Figure 2 . In all categories over 85% of trials included response outcomes, besides the trials with unknown status (77.8%). In terms of MRD, almost 50% of trials with recruiting (46.0%) or not yet recruiting (46.8%) status had MRD as an outcome. However, in case of completed trials, this value was only 16.9%, while it was 12.0% for suspended, terminated or withdrawn trials.

Proportion of outcomes by phase of the trial is presented on Figure 3 . In all categories over 85% of trials included response outcomes, besides the Phase 4 trials (60.0%) and those with not available data on the phase (58.1%). In terms of MRD, 61.1% of Phase 3 trials had MRD as an outcome. This proportion was also fairly high in Phase 2 trials (45.3%), while it was lower in Phase 4 (6.7%) and Phase 1 trials (19.9%).

In total, there were 106 trials, where MRD was listed as a primary outcome, which was 28.4% of those trials, where MRD was assessed. Details by year, trial status and phase of the trial are presented in the tables of Supporting Information : Appendix 2 . While in 2014, there was no trial with MRD as a primary outcome, in 2025 there were 20, which was 47.6% of the total trials with MRD in that year. Since 2018, the proportion of trials with MRD as a primary outcome was at least 20% in every year among trials with MRD. In terms of trial status, for completed trials, there were only 2 with MRD as a primary outcome in our dataset, which was 4.4% of the total completed trials with MRD. The highest proportion was observed for trials with recruiting (36.6%) and not yet recruiting (32.6%) status. By phase of the trial, the proportion of MRD as a primary outcome showed that it was 26.1% for Phase 3, 39.7% for Phase 2 and 9.1% in Phase 1 trials (among those that had MRD assessment).

In total, there were 28 trials, where MRD was used to guide therapeutic decisions, which was 7.5% of those trials, where MRD was assessed. Details of these trials by year, trial status and phase of the trial are presented in the tables of Supporting Information : Appendix 3 . The highest number of trials, where MRD was used to guide therapy, was in 2023 with 8. There were 3 trials, which were already completed, while 18 were in the recruiting stage and 7 in not yet recruiting. The majority of trials, where MRD was used to guide therapy, were Phase 2 ( n = 18), while Phase 3 ( n = 4) and phase 1 ( n = 1) were less common.

We further investigated the type of treatment among those trials, where MRD was measured by specifically focusing on innovative therapies, CAR‐T, and bi‐specific antibodies ( Supporting Information : Appendix 4 ). We identified 66 trials with CAR‐T (17.7%), among which 3 were completed, 33 were recruiting, 17 were active but not recruiting, 2 terminated, and 11 unknown trials. Most studies were Phase 2 ( n = 24), while there were 7 Phase 3 trials. We identified 85 trials with bi‐specific antibodies (22.8%), among which 4 were completed, 47 were recruiting, 31 were active but not recruiting, 2 terminated/withdrawn, and 1 unknown trials. Most studies were Phase 2 ( n = 38), while there were 14 Phase 3 trials.

In terms of the targeted patient population, we identified 114 trials focusing on newly diagnosed (30.6%) and 180 on relapsed/refractory patients (48.3%), while this information was clearly identifiable for the remaining 21.2% of the trials with MRD assessment (see Supporting Information : Appendix 5 ). Newly diagnosed patients were further categorized as transplant eligible ( n = 53), ineligible ( n = 22), and not clearly identifiable in terms of transplant eligibility ( n = 39).

Details on the method of MRD assessment were limited in many cases. The technology used for the measurement was reported only in 169 trials in the database, which is 45.3% of the total trials with MRD assessment. Among these, 91 used flow cytometry approaches including NGF and 104 NGS, while 26 used both. Details on the technologies by year are presented in the table of Supporting Information : Appendix 6 . No clear trend was possible to detect over time, which might be attributed to reporting bias in the original database. However, the first trial that used both types of technologies was already found in 2014, and by 2023 and 2024, there were seven such trials in each year. The use of PET/CT imaging for the MRD assessment was rarely mentioned specifically among the reviewed trials ( n = 9).

Details on the cut‐off for MRD negativity were also collected and the details are presented in Supporting Information : Appendix 7 . A total of 130 studies provided such information, which is 34.9% of the trials with MRD assessment. A single cut‐off at 10 −5 was most frequently used. A total of 28 studies used at least a 10 −6 cut‐off, and this has been increasingly used in 2024 and 2025, with 7 and 6 studies, respectively. In addition, we also investigated if trials evaluated MRD according to the IMWG criteria. Out of the 373 trials with MRD assessment, this information was reported only for 78 cases (20.9%). This proportion was similar for trials with MRD as a primary outcome (24.1%). However, it is important to note that some outcome descriptions were very short in the database.

Finally, we attempted to investigate the timing of MRD assessment in the clinical trials. In total, 150 reported such information, which is 40.2% of the trials with MRD assessment. Unfortunately, this data was not standardized across trials, and it was not possible to detect key trends on this parameter. Rather, we categorized trials with assessment at a single time ( n = 37) and trials with assessment at multiple timepoints ( n = 103).

Apart from the interventional clinical trials, we also identified 127 observational studies in the database. Therapy response was assessed in 55.9% ( n = 71), while MRD was investigated in 14.2% ( n = 18). Among these latter ones, 2 were completed, 14 were recruiting, and 2 were not yet recruiting. The first observational study with MRD was initiated in 2018, while the most was identified in 2024 with six studies. MRD was a primary outcome in seven observational studies, and it was used to guide therapeutic decisions in five cases. Type of MRD assessment was available only in six cases; flow cytometry‐based was used in four, NGS in five, while both technologies were used in three studies.

Discussion of Study Findings

Our timely review of MM trial outcomes showed a clearly increasing trend of MRD assessment and the observation of almost 50% of studies with recruiting or not yet recruiting status indicates that a vast amount of evidence will be generated in the future on various therapeutic alternatives for this outcome. This highlights an important need for strong coordination of MRD evidence generation, and joint collaborative work with drug regulatory agencies [ 14 ]. Without deliberate effort in these areas, the field may continue to fall short of transforming MRD from a promising biomarker into a fully accepted surrogate endpoint for drug approval.

Many of our findings are directly comparable with a former study published in 2022 that had a similar overview with a shorter, 5‐year timeframe between 2015 and 2020 [ 13 ]. They reviewed much fewer trials ( n = 598), compared to our database with 1209. In their time period, the MRD was assessed in 24.2% of the trials, while our figure was 30.9%. They found that MRD as a primary endpoint was included in 24.8% of trials with MRD, while our figure was similar, 28.4%. They showed that 6.2% utilized MRD assessment as a stratification tool to determine treatment, and our figure was also similar, 7.5%. Specific years are also directly comparable; their lowest figure in terms of trials with MRD assessment was in 2015 (13.3%, similar to our figure in the same year, which was 15.3%), while their highest figure was in 2019 (33.1%, while our figure was 27.6%). Finally, they also assessed the technology used for MRD, which was found in 54.4% of the studies with MRD (our figure was 45.3%). While there are slight differences, which can be explained by the different time periods and by the application of research methods, the former study provides an important face validity for our findings.

MM exemplifies the therapeutic paradigm that deeper responses are associated with longer survival. This principle is supported by both meta‐analyses of clinical trials and observational studies from routine clinical practice, which consistently show that achieving a complete remission is associated with improved progression‐free and OS [ 15 ]. Importantly, complete remission is a prerequisite for most definitions of MRD negativity, whether assessed by NGH, NGS, or imaging modalities such as PET/CT. Furthermore, the concept of sustained MRD negativity, which requires confirmation of complete remission over time, has emerged as an even more powerful predictor of long‐term outcomes. As a result, MRD builds upon and refines traditional response criteria, offering a more sensitive and dynamic framework for evaluating treatment efficacy in both clinical trials and practice.

The recent advice issued by the EMA in November 2024 marks a pivotal step toward accepting MRD as a regulatory endpoint in MM trials (see detailed in Supporting Information : Appendix 1 ). While CHMP acknowledged the prognostic value of MRD negativity—particularly at thresholds of 10 −5 or higher—and its potential to support conditional approvals in specific settings, they also highlighted critical gaps that must be addressed. Chief among these is the lack of assay standardization and uncertainties around generalizability across treatment modalities. These regulatory signals underscore the importance of generating high‐quality, consistent evidence that links MRD to long‐term outcomes in well‐defined clinical contexts [ 16 ].

From a health technology assessment (HTA) perspective, the acceptability of MRD as a surrogate endpoint introduces additional complexity. Unlike regulatory agencies, which may grant conditional approval based on intermediate endpoints, HTA bodies often require mature data on OS, quality of life, and cost‐effectiveness to support reimbursement decisions [ 17 ]. Although the prognostic value of MRD negativity is increasingly recognized, its qualification as a surrogate for long‐term clinical benefit has not yet reached consensus among HTA agencies [ 18 ]. This presents a challenge for market access, particularly in cases where MRD‐driven approvals precede definitive survival data. For MRD to be accepted in HTA evaluations, further evidence and dialogues will be needed to establish consistent, quantifiable links between MRD status and patient‐relevant outcomes across diverse treatment strategies and clinical settings. As such, the integration of MRD into reimbursement pathways will likely depend not only on methodological standardization but also on robust validation of its predictive value in real‐world and late‐phase clinical data.

In parallel with regulatory and HTA considerations, EMA's Reflection Paper on Patient Experience Data underscores the growing importance of integrating patient perspectives into evidence generation and regulatory evaluation [ 19 ]. The paper highlights that patient‐derived data, such as quality of life measures, treatment preferences, and perceived disease burden, are essential for contextualizing clinical and biomarker outcomes, ensuring that regulatory decisions reflect patient priorities. In the context of MRD, the incorporation of patient experience data could strengthen the interpretability and relevance of MRD‐driven endpoints by linking biological response to outcomes that matter most to patients, such as functional improvement and treatment tolerability. This approach aligns with EMA's broader move toward patient‐focused benefit–risk assessment and suggests that future MRD research should not only meet technical and methodological standards, but also embed patient experience as a complementary dimension of clinical value. Such integration would facilitate a more holistic evidence framework that supports both regulatory approval and downstream HTA decision‐making.

Our study has important limitations. First of all, there have been many studies from other disease areas as well that relied on data from the ClinicalTrials.gov database. It is important to acknowledge the limitations of these types of studies, mainly due to the changes over time in the data collected, the definitions used, and the rigor with which missing data was pursued [ 20 ]. These limitations are especially true for reporting trial outcomes, compared to reviews of peer reviewed publications [ 21 ]. On the other hand, the database is widely known in clinical research, its usability is simple, and it is applicable for identifying major trends and observing patterns, especially on information related to trial design. Even though trial protocols are richer and more valuable sources of trial design information, their identification and detailed analysis face other barriers such as limited availability from public sources. Our findings could have been compared to protocol‐based information collection; however, within the scope of this research we had no opportunity to conduct such a document search process.

Key Areas of Current Research Agenda

Define optimal timing for MRD assessment in clinical trials: Many trials lack standardized MRD assessment timepoints. Further research should determine when MRD testing provides the greatest prognostic and therapeutic value across treatment lines and settings.

Clarify the role of MRD as a treatment‐guiding biomarker: Despite increasing use of MRD to guide therapy, evidence is limited on the outcomes of MRD‐adapted treatment strategies. Prospective studies should evaluate how MRD‐driven decisions affect long‐term survival, quality of life, and treatment burden.

Improve transparency and consistency in trial registry reporting: Our study identified incomplete or inconsistent reporting of MRD methodology, cut‐off values, and outcome categorization. Efforts are needed to enhance the quality of trial registry data to support secondary research and regulatory analyses.

Explore the intersection of MRD and patient‐relevant endpoints: While MRD is biologically meaningful, its correlation with patient‐reported outcomes, functional status, and life expectancy in real‐world populations remains underexplored.

Align trial design with evolving regulatory and HTA expectations: With MRD increasingly recognized by regulators, future studies should be prospectively designed to support both regulatory and reimbursement decision‐making, including integration of complementary endpoints such as PFS and health‐related quality of life.

Future Considerations

Developments in MRD assessment aim to expand both the sensitivity and accessibility of testing, with particular focus on non‐invasive and real‐time monitoring strategies. While bone marrow remains the current gold standard for MRD analysis, its limitations—including invasiveness, sampling bias, and spatial heterogeneity—are driving innovation toward peripheral blood‐based assays. Techniques such as mass spectrometry, circulating tumor DNA (ctDNA) profiling, and single‐cell sequencing of plasma cells from blood are under investigation and have shown promising early results in detecting residual disease with high concordance to marrow‐based methods, especially when combined with immunophenotypic or genomic markers [ 22 , 23 , 24 ]. These approaches may ultimately enable more frequent, less invasive MRD monitoring that could better reflect systemic disease burden and support adaptive treatment decisions over time [ 25 ]. In addition, systematic screening for precursor conditions such as monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (SMM) can improve early detection, as recently demonstrated in the iStopMM study [ 26 , 27 ].

Technological advances are also extending into imaging and functional diagnostics. Integration of PET/CT with molecular MRD testing continues to improve risk stratification and capture disease outside the marrow compartment, but next‐generation imaging tools and radiotracers are being developed to enhance sensitivity and specificity even further [ 24 ]. Artificial intelligence and machine learning applications are also beginning to emerge for automated image interpretation and predictive modeling of MRD dynamics [ 28 ]. The use of monoclonal antibodies to generate artificial MRD reference samples could support the validation of MRD assays across various technologies and thresholds, improving reproducibility and cross‐platform standardization. As these technologies mature, their incorporation into clinical workflows will depend not only on analytical validity but also on clinical utility, demonstrating that MRD‐guided treatment changes lead to meaningful improvements in patient outcomes [ 29 ]. Continued prospective studies and harmonization of assay platforms will be essential to bridge the gap between innovation and implementation in routine care.

Conclusions

This comprehensive analysis demonstrates the rapid evolution of MRD from a prognostic biomarker to a potential regulatory and clinical endpoint. The increasing proportion of trials incorporating MRD, particularly as a primary or adaptive endpoint, reflects a paradigm shift toward deeper response assessment and individualized treatment evaluation. Nevertheless, our findings underscore persistent heterogeneity in MRD methodology, timing, and reporting, which continues to limit comparability across studies. The recent qualification advice from the EMA represents a critical regulatory milestone, recognizing MRD negativity as a valid prognostic measure and a candidate surrogate endpoint, while emphasizing the need for methodological standardization and demonstration of long‐term benefit. To achieve broader regulatory and HTA acceptance, MRD research must align with both evidentiary and patient‐centered standards, integrating outcomes that capture quality of life and treatment impact.

Author Contributions

M.C.Q., G.D.O., A.I., and M.C. designed the study. B.A.‐A., S.K., A.M.T., I.M., and A.P.O. collected the data. J.‐L.H., A.S., D.H., and E.S. provided input for interpretation. All authors approved the final version of the manuscript.

Funding

Syreon Research Institute received research grants through Building Medicine Bridges in Cancer Care (BMB 2.0) Consortium at Patients for Partners NGO to partner on this study.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Appendix 1: EMA‐CHMP's Qualification advice on the use of MRDnegCR as an intermediate early endpoint for conditional market approval in myeloma clinical trials.

Appendix 2: Overview of trials, where MRD was a primary outcome.

Appendix 3: Overview of trials, where MRD was used to guide therapeutic decisions.

Appendix 4: Details on trials focusing on CAR‐T therapy or bi‐specific antibodies.

Appendix 5: Details on trials depending on their patient population.

Appendix 6: Details on the type of technology used for MRD assessment.

Appendix 7: Details on the cut‐off level of MRD assessment.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

1 R. Szalat , K. Anderson , and N. Munshi , “ Role of Minimal Residual Disease Assessment in Multiple Myeloma ,” Haematologica 109 , no. 7 ( 2024 ): 2049 – 2059 , 10.3324/haematol.2023.284662 . 38328864 PMC11215375

2 H. Quach , “ MRD End Point in Myeloma: Ready for Prime Time? ,” Blood 139 , no. 6 ( 2022 ): 799 – 802 . 35142848 10.1182/blood.2021013363

3 N. C. Munshi , H. Avet‐Loiseau , K. C. Anderson , et al., “ A Large Meta‐Analysis Establishes the Role of MRD Negativity in Long‐Term Survival Outcomes in Patients With Multiple Myeloma ,” Blood Advances 4 , no. 23 ( 2020 ): 5988 – 5999 . 33284948 10.1182/bloodadvances.2020002827 PMC7724898

4 S. Kumar , B. Paiva , K. C. Anderson , et al., “ International Myeloma Working Group Consensus Criteria for Response and Minimal Residual Disease Assessment in Multiple Myeloma ,” Lancet Oncology 17 , no. 8 ( 2016 ): e328 – e346 . 27511158 10.1016/S1470-2045(16)30206-6

5 E. E. Manasanch , “ What to Do With Minimal Residual Disease Testing in Myeloma ,” Hematology. American Society of Hematology. Education Program 2019 , no. 1 ( 2019 ): 137 – 141 . 31808833 10.1182/hematology.2019000080 PMC6913484

6 Q. Shi , B. Paiva , L. D. Pederson , et al., “ Minimal Residual Disease‐Based End Point for Accelerated Assessment of Clinical Trials in Multiple Myeloma: A Pooled Analysis of Individual Patient Data From Multiple Randomized Trials ,” Journal of Clinical Oncology 43 , no. 11 ( 2025 ): 1289 – 1301 . 39938021 10.1200/JCO-24-02020

7 A. Perrot , J. Lambert , C. Hulin , et al., “ Measurable Residual Disease‐Guided Therapy in Newly Diagnosed Myeloma ,” New England Journal of Medicine 393 , no. 5 ( 2025 ): 425 – 437 . 40459097 10.1056/NEJMoa2505133 PMC13292240

8 L. Bertamini , M. D'Agostino , and F. Gay , “ MRD Assessment in Multiple Myeloma: Progress and Challenges ,” Current Hematologic Malignancy Reports 16 , no. 2 ( 2021 ): 162 – 171 . 33950462 10.1007/s11899-021-00633-5

9 R. Fonseca , M. Arribas , J. E. Wiedmeier‐Nutor , et al., “ Integrated Analysis of Next Generation Sequencing Minimal Residual Disease (MRD) and PET Scan in Transplant Eligible Myeloma Patients ,” Blood Cancer Journal 13 , no. 1 ( 2023 ): 32 . 36878906 10.1038/s41408-023-00794-x PMC9988896

10 F. E. Lecouvet , M. C. Vekemans , T. Van Den Berghe , et al., “ Imaging of Treatment Response and Minimal Residual Disease in Multiple Myeloma: State of the Art WB‐MRI and PET/CT ,” Skeletal Radiology 51 , no. 1 ( 2022 ): 59 – 80 . 34363522 10.1007/s00256-021-03841-5 PMC8626399

11 B. Paiva , Q. Shi , N. Puig , et al., “ Opportunities and Challenges for MRD Assessment in the Clinical Management of Multiple Myeloma ,” Nature Reviews Clinical Oncology 22 , no. 6 ( 2025 ): 424 – 438 . 10.1038/s41571-025-01017-x 40195455

12 “Access to Corresponding FDA Transcript,” accessed 1st October 2025, https://www.fda.gov/media/180109/download .

13 O. Van Oekelen , N. Birrer , W. Wesson , et al., “ Characteristics of Measurable Residual Disease Assessment in Myeloma: A Review of Clinical Trials From 2015‐2020 ,” Blood Cancer Journal 12 , no. 11 ( 2022 ): 155 . 36379917 10.1038/s41408-022-00750-1 PMC9666639

14 O. Landgren and S. M. Devlin , “ Minimal Residual Disease as an Early Endpoint for Accelerated Drug Approval in Myeloma: A Roadmap ,” Blood Cancer Discovery 6 , no. 1 ( 2025 ): 13 – 22 . 39630969 10.1158/2643-3230.BCD-24-0292 PMC11707509

15 B. Paiva , J. San‐Miguel , and H. Avet‐Loiseau , “ MRD in Multiple Myeloma: Does CR Really Matter? ,” Blood 140 , no. 23 ( 2022 ): 2423 – 2428 . 35560160 10.1182/blood.2022016170

16 M. V. Mateos and N. Puig , “ Accelerating Myeloma Drug Development: Will Minimal Residual Disease Replace Progression‐Free Survival as an Intermediate Surrogate End Point? ,” Journal of Clinical Oncology 43 , no. 11 ( 2025 ): 1271 – 1274 . 40048684 10.1200/JCO-25-00048

17 M. Boyiadzis , A. H. Wei , B. Paiva , et al., “ Measurable Residual Disease (MRD) as a Surrogate End Point for Clinical Drug Approval in Acute Myeloid Leukemia (AML): Perspectives From the MRD Partnership and Alliance in AML Clinical Treatment Consortium ,” Cancer 131 , no. 13 ( 2025 ): e35960 . 40576165 10.1002/cncr.35960 PMC12203635

18 W. Grandy , P. Kachouh , and B. Yang , “ POSC284 the Acceptance of Minimal Residual Disease (MRD) as a Clinical Endpoint in Multiple Myeloma (MM): A Targeted Literature Review ,” Value in Health 25 , no. 1 ( 2022 ): S189 – S190 .

19 “Reflection Paper on Patient Experience Data,” accessed 1st October 2025, https://www.ema.europa.eu/en/documents/scientific‐guideline/reflection‐paper‐patient‐experience‐data_en.pdf .

20 R. M. Califf , D. A. Zarin , J. M. Kramer , R. E. Sherman , L. H. Aberle , and A. Tasneem , “ Characteristics of Clinical Trials Registered in ClinicalTrials.Gov, 2007–2010 ,” Journal of the American Medical Association 307 , no. 17 ( 2012 ): 1838 – 1847 . 22550198 10.1001/jama.2012.3424 PMC13431740

21 D. M. Hartung , D. A. Zarin , J. M. Guise , M. McDonagh , R. Paynter , and M. Helfand , “ Reporting Discrepancies Between the ClinicalTrials.Gov Results Database and Peer‐Reviewed Publications ,” Annals of Internal Medicine 160 , no. 7 ( 2014 ): 477 – 483 . 24687070 10.7326/M13-0480 PMC4617780

22 B. Paiva and J. F. San‐Miguel , “ Myeloma: A Sky Full of MRD Stars ,” Blood 146 , no. 6 ( 2025 ): 649 – 650 , 10.1182/blood.2025029160 . 40773178

23 N. S. Chandhok and M. A. Sekeres , “ Measurable Residual Disease in Hematologic Malignancies: A Biomarker in Search of a Standard ,” EClinicalMedicine 86 ( 2025 ): 103348 , 10.1016/j.eclinm.2025.103348 . 40666170 PMC12257026

24 A. Medina‐Herrera , M. E. Sarasquete , C. Jiménez , N. Puig , and R. García‐Sanz , “ Minimal Residual Disease in Multiple Myeloma: Past, Present, and Future ,” Cancers (Basel) 15 , no. 14 ( 2023 ): 3687 , 10.3390/cancers15143687 . 37509348 PMC10377959

25 T. Kubicki , B. A. Derman , D. Dytfeld , and A. J. Jakubowiak , “ Measurable Residual Disease in Peripheral Blood in Myeloma: Dream or Reality ,” Current Opinion in Oncology 35 , no. 6 ( 2023 ): 574 – 580 . 37621165 10.1097/CCO.0000000000000987

26 A. Ý. Sigurbergsdóttir , S. Rögnvaldsson , S. Thorsteinsdóttir , et al., “ Disease Associations With Monoclonal Gammopathy of Undetermined Significance Can Only Be Evaluated Using Screened Cohorts: Results From the Population‐Based iStopMM Study ,” Haematologica 108 , no. 12 ( 2023 ): 3392 – 3398 . 37439374 10.3324/haematol.2023.283191 PMC10690899

27 S. Thorsteinsdóttir , G. K. Gíslason , T. Aspelund , et al., “ Prevalence of Smoldering Multiple Myeloma Based on Nationwide Screening ,” Nature Medicine 29 , no. 2 ( 2023 ): 467 – 472 , 10.1038/s41591-022-02183-6 . PMC11099842 36747117

28 N. Radakovich , M. Nagy , and A. Nazha , “ Machine Learning in Haematological Malignancies ,” Lancet Haematology 7 , no. 7 ( 2020 ): e541 – e550 . 32589980 10.1016/S2352-3026(20)30121-6

29 L. N. Gordan , A. Warner , T. Heritage , et al., “ Community Hematologist/Oncologist Perspectives on Minimal Residual Disease Testing in Patients With Multiple Myeloma ,” Future Oncology 21 , no. 22 ( 2025 ): 2875 – 2883 . 40832984 10.1080/14796694.2025.2546284 PMC12439568