Is MRD Testing Worthwhile in Myeloma?

Faith Davies, MD
Professor of Medicine, Myeloma Institute for Research and Therapy, Winthrop P. Rockefeller Cancer Institute, University of Arkansas for Medical Sciences
Saad Z. Usmani, MD
Director of Plasma Cell Disorders Program, Director of Clinical Research in Hematologic Malignancies, Levine Cancer Institute

With the clinically effective treatment regimens now available for myeloma, more patients can achieve deep and sustained responses. The ability to measure minimal residual disease (MRD) during and after treatment has the potential to help guide treatment decisions and predict patients’ likelihood of relapse, but is MRD monitoring ready to be used in routine clinical practice?

ASH Clinical News invited Faith Davies, MD, and Saad Z. Usmani, MD, to debate the question, “Is measuring MRD worthwhile in myeloma?” Dr. Davies, professor of medicine and deputy director of the Myeloma Institute at the University of Arkansas for Medical Sciences, will argue on the “pro” side; Dr. Usmani, chief of the Plasma Cell Disorders Program and director of clinical research in hematologic malignancies at Levine Cancer Institute/Carolinas HealthCare System, will argue on the “con” side.

Faith Davies, MD: Recently, a number of new myeloma treatment options have become available, with four new agents receiving U.S. Food and Drug Administration (FDA) approval since 2015: the monoclonal antibodies daratumumab and elotuzumab, the histone deacetylase inhibitor panobinostat, and the proteasome inhibitor ixazomib. This proliferation of options has helped more patients with myeloma achieve a complete response (CR) and, along with these new treatment options, the technology for detecting low levels of disease has also improved to the point that we are able to monitor patients’ MRD effectively.

Measuring MRD is informative because patients who achieve an MRD-negative state have a much better prognosis. In a recent meta-analysis of 21 studies that assessed MRD status and progression-free survival (PFS) and overall survival (OS) in patients with newly diagnosed myeloma, the authors found that patients who achieved MRD-negative status had a PFS nearly twice as long as patients who were MRD-positive (54 vs. 26 months; p<0.001).1 MRD-negative patients also had longer OS (98 vs. 82 months; p<0.001), and the association between MRD status and survival persisted among patients who achieved CR. There’s no doubt that it is worthwhile for clinicians to measure MRD as a signal of how well patients are responding to treatment and to serve as a guide as to their overall prognosis.

Saad Z. Usmani, MD: In principle, yes, MRD testing is a fantastic idea. My concern is that we haven’t fully vetted the available platforms. Even among patients who achieve CR, we don’t know the true measure of underlying residual disease because we still haven’t figured out the optimal technique for collecting these data.

The most common method for detecting MRD in bone marrow (BM) is flow cytometry. However, with this method, we can only obtain a finite number of samples, and the depth of detectable MRD negativity is not optimal. The estimated sensitivity is only 10-4; other methods allow us to detect myeloma cells down to 10-6.

There is a lot of interest in developing next-generation sequencing (NGS) techniques to measure MRD, with the hope that they will overcome some of the limitations of flow cytometry and polymerase chain reaction assays. This technique isn’t ready to be used routinely in clinical practice and the company producing the only commercially available NGS testing kit is trying to keep pace with the growing sample volumes. And these are just the samples being analyzed from clinical trials, leaving us to wonder what will happen when that workload grows as physicians start incorporating NGS techniques more regularly in their practices.

These are the two methodologies of greatest interest right now, but companies are also actively developing other platforms. There is no clear consensus on which methodology to use or when to use it.

Regardless of how the samples are analyzed, all methods rely on BM aspirates taken from a single site – typically, the pelvic bone – ignoring the fact that patients can have disease in many other areas of the body. Measuring disease burden in just one area may not give us an accurate picture of the patient’s disease.

Dr. Davies: There are a number of tools in development to measure MRD and it seems unlikely that there is just one ideal test. In the long run, we may find that all of the tools are valid and provide clinically useful information. The most important factor to consider is the test’s sensitivity. Available tests are able to document tumor cells at levels of 10‑4 to 10‑6.

I think your concerns about the availability of MRD testing on a large scale are valid. MRD testing has become more commonplace in the setting of acute leukemias, but it has only really been used in the clinical trial setting in myeloma to date. I’m certain that transplant centers conduct MRD monitoring with more regularity.

Dr. Usmani: Absolutely. Even though the International Myeloma Working Group (IMWG) issued consensus criteria defining the role of MRD assessment in 2016, many hospitals are probably still picking up the tab for the procedure.2

“At this moment, we don’t have any data to support what defines ‘sustained’ minimal residual disease negativity or at what time point we could consider scaling back treatment.”
—Faith Davies, MD

Regardless, before MRD monitoring becomes part of routine clinical practice, we need to ensure that MRD is being measured and validated uniformly – whether that’s by flow cytometry or sequencing.

Unfortunately, there isn’t much research comparing MRD testing with other efficacy endpoints head-to-head in different patient settings. There are two large datasets that have looked prospectively at the value of MRD in patients with relapsed myeloma: CASTOR and POLLUX, both of which served as the basis for the FDA’s approval of daratumumab for patients with relapsed/refractory myeloma.3 In those clinical trials, we observed that even high-risk patients were able to achieve MRD negativity when treated with daratumumab (in combination with either lenalidomide and dexamethasone or bortezomib and dexamethasone) and, in general, MRD-negative patients had longer PFS in both of those studies compared with MRD-positive patients.

Beyond that, we do not have enough prospective trial data to comment on the value of MRD testing in high-risk patients.

Dr. Davies: I am hopeful that we will have more answers to these questions, particularly concerning high-risk myeloma, when we see results from two research groups that are investigating the use of flow cytometry MRD testing in the upfront setting: GEM-PETHEMA (the Spanish Myeloma Group) and the U.K. Myeloma Research Group. A pooled analysis of studies conducted by the Spanish group determined that patients who achieved CR but remained MRD positive had a similar PFS, compared with those who achieved PR (27 vs. 29 months; p<0.001).4 Achieving MRD negativity, however, was significantly associated with prolonged PFS (63 months; p<0.001), suggesting that MRD status had greater predictive significance than CR status alone. The researchers also noted that this association was particularly strong in patients with high-risk cytogenetics.

Results from the U.K. group are forthcoming.

On the other hand, a recent study of 185 patients with myeloma showed that, although MRD negativity was associated with longer PFS and OS than MRD positivity, MRD negativity did not translate to superior survival in patients with del17p or ≥2 cytogenetic abnormalities, suggesting that MRD testing may not be valuable in patients with these risk factors because they have poor outcomes regardless of MRD status.5

We are also eagerly anticipating updated results from the MRD analysis of the Intergroupe Francophone du Myelome/Dana-Farber Cancer Institute 2009 study, which is measuring MRD with both flow cytometry and NGS techniques (which can detect myeloma cells to a level of 10-4 and 10-6, respectively) in newly diagnosed patients, including those with high-risk disease.

Data suggest that achieving MRD negativity is predictive of survival. These results have prompted researchers to question whether MRD negativity could be used as a surrogate endpoint for survival in drug approvals. However, I don’t think we’re quite ready for that.

Dr. Usmani: I am not prepared to say that MRD status should be used as a surrogate marker either, and I think we should be cautious about using MRD status to drive any treatment decisions given the concerns we discussed earlier – namely a lack of guidance on what constitutes MRD and a lack of standardization in analyzing MRD samples.

The IMWG consensus guidelines established an acceptable sensitivity level of 10-5, but I think many hematologists tend to agree that 10-6 is an ideal MRD negativity sensitivity point.2

Dr. Davies: The guidelines provide sensitivity cut-points for both flow cytometry and NGS. These serve as baselines so that investigators can start to report in a more standardized fashion and compare results between different treatment strategies. As MRD monitoring is used more frequently in practice, the IMWG guidelines will need to be updated accordingly, and it will become clear what level of sensitivity is most appropriate for each situation.

In my opinion, the prerequisite for myeloma to be considered “cured” is to achieve sustained MRD negativity. We need to be using the best treatments possible for every patient, but, even if a patient reaches an MRD-negative status, it doesn’t necessarily mean treatment should be stopped. I would argue that MRD positivity should play a greater role in making treatment decisions; if a patient reaches MRD positivity after treatment, that’s when we need to stop and think, “Do we change our treatment plan to improve the depth of response?”

At the moment, though, we don’t have any data to support what defines “sustained” MRD negativity or at what time point we could consider scaling back treatment. The next series of clinical trials, which many of the cooperative groups are discussing and trying to design, are hopefully going to provide some answers to these questions.

The studies about MRD-guided therapy that we have now are largely retrospective trials; we need prospective trials to provide guidance. Personally, I would need a patient to be testing MRD negative for at least 1 year, but more likely 3 years, before I would consider de-escalating treatment.

Dr. Usmani: I agree – at least 1 year, but ideally 3 years of sustained MRD negativity before you can say a patient is potentially cured or is going to survive long term.

And, before we can start to develop response-adaptive treatment strategies based on MRD status – either de-escalating treatment if the patient is MRD negative or changing the treatment plan if the patient is MRD positive – we need to better define what “counts” as MRD negativity. Sustained MRD negativity should be our treatment goal, but this is an issue that clinical trials are just now starting to address.


  1. Munshi NC, Avet-Loiseau H, Rawstron AC, et al. Association of minimal residual disease with superior survival outcomes in patients with multiple myeloma: a meta-analysis. JAMA Oncol. 2017;3:28-35.
  2. Kumar S, Palva B, Anderson KC, et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 2016;17:e328-46.
  3. Avet-Loiseau H, Casneuf T, Chiu C, et al. Evaluation of minimal residual disease (MRD) in relapsed/refractory multiple myeloma (RRMM) patients treated with daratumumab in combination with lenalidomide plus dexamethasone or bortezomib plus dexamethasone. Abstract #246. Presented at the 2016 ASH Annual Meeting, December 3, 2016; San Diego, California.
  4. Lahuerta J, Palva B, Vidriales M, et al. Depth of response in multiple myeloma: a pooled analysis of three PETHEMA/GEM clinical trials. J Clin Oncol. 2017 May 12. [Epub ahead of print]
  5. Chakraborty R, Muchtar E, Kumar SK, et al. Impact of post-transplant response and minimal residual disease on survival in myeloma with high-risk cytogenetics. Biol Blood Marrow Transplant. 2017;23:598-605.

Disclaimer: The following positions were assigned to the participants and do not necessarily reflect ASH’s opinion, the participants’ opinions, or what they do in daily practice.

Agree? Disagree? We want to hear from you! Send your thoughts and opinions on this controversial issue to