Minimal residual disease (MRD) negativity can be achieved and has clear prognostic value in multiple myeloma. However, efforts to understand how best to use MRD testing clinically are still ongoing.

Getting fewer than 1 in 100,000 or 1 in 1,000,000 multiple myeloma cells in the bone marrow after treatment is very exciting. In the past, we have only been able to achieve that with allogeneic transplantation, but now we can achieve MRD negativity at a high frequency with the advent of novel treatments. A large meta-analysis showed that MRD negativity confers prolonged progression-free survival (PFS) and overall survival (OS) advantages in patients with newly diagnosed multiple myeloma or in those with relapsed/refractory multiple myeloma.

The biologic excitement is that if you can get to MRD negativity and get rid of the abnormal multiple myeloma clone early, you can potentially prevent the evolution that inevitably occurs in multiple myeloma and leads to disease resistance and, ultimately, relapse. The achievement of MRD negativity is no longer a dream, which is exciting, but that is not enough. Rather, the durability of MRD negativity is also important. Persistent MRD negativity is associated with a long-term benefit after treatment in newly diagnosed, transplant-eligible and -ineligible patients, as well as after chimeric antigen receptor T-cell therapy for multiply relapsed disease.

Another potential application of MRD analysis that is being studied is in the context of using MRD to guide post-transplant consolidation and possible treatment cessation. The current paradigm is lifelong maintenance therapy. Data from the phase 2 MASTER trial showed that the 2-year cumulative risk for progression for patients with 0, 1, or 2 or more high-risk cytogenetic abnormalities was 9%, 9%, and 47%, respectively. Such results demonstrate that, while we are excited that we can achieve MRD negativity, we are still learning how to use MRD negativity clinically.

Another hope with MRD negativity is that it can be used as a regulatory end point in appropriately designed clinical trials. Unprecedented PFS is now being observed in newly diagnosed and relapsed/refractory multiple myeloma, and, unless alternative end points to PFS are identified, new studies will not have a readout for many years, which may be too late for many of our patients.

One final thought is to emphasize the importance of assessing MRD negativity inside and outside the bone marrow. We need to be sure that we also have imaging negativity (eg, positron emission tomography/computed tomography or magnetic resonance imaging) when we assess MRD negativity, as defined in the International Myeloma Working Group criteria.

MRD cutoffs for prognosis have become clearer, and it seems that, if prognostic information is the goal, then 10^-6 seems much better for predicting outcomes than 10^-5. It is also important to emphasize that a single result of MRD negativity has very little meaning compared with a sustained MRD negativity report across multiple time points. If we are going to use MRD to guide treatment decisions such as treatment discontinuation, we need to have multiple MRD-negative readouts because that has more prognostic power than a single reading.

We are beginning to explore new ways to estimate MRD. Next-generation sequencing and next-generation flow cytometry are methods that are used with bone marrow samples, but there is also increasing interest in alternative methods that might allow us to reduce the need for more frequent bone marrow biopsies. In that regard, very sensitive mass spectrometry or other genomic methods might be used for MRD testing. Interesting results have been reported with mass spectrometry–based methodologies to predict survival outcomes, but further investigation is needed.

It is difficult to use early MRD assessments at 4 and 6 months to determine long-term outcomes. I think that the focus is appropriately on MRD in high-risk patients in the sense that we are really keen on getting high-risk patients to an MRD-negative status. The higher the patient’s risk, the more we want to find a way to modify therapy to get them to MRD negativity. While this is not fully data driven, it is based on our philosophy that these patients do not do well unless they become MRD negative.

There are many studies in which patients who achieve MRD negativity have a longer PFS and OS than patients who do not. And it is absolutely true that achieving such a low disease burden correlates with longer PFS, much like achieving a complete response correlated with better survival 10 years ago. However, we also know that many patients who achieve MRD negativity will ultimately relapse. In many cases, MRD negativity may give us more information about the biology and responsiveness of the disease to a given therapy than the curative potential of the therapy.

Thus, I see MRD as another variable that we need to understand. To even consider discontinuation, I think that we would need to use 10^-6 as the cutoff and to see MRD negativity sustained for 12 months or longer. For patients with multiple myeloma who are at standard risk and receive lenalidomide, bortezomib, and dexamethasone in combination with autologous stem cell transplantation followed by lenalidomide maintenance, the estimated median PFS is 76.5 months. So, short-term end points such as MRD testing at cycle 4 or 8 pale in comparison to the follow-ups after 50 months that we see with the current continuous therapy model.