NEW YORK, NY—Speakers faced off over the issue of minimal residual disease (MRD) testing in multiple myeloma (MM) at Lymphoma & Myeloma 2017.
Ola Landgren, MD, PhD, of Weill Cornell Medicine in New York, New York, said, “it’s really a necessary and logical step forward to look at MRD.”
On the other hand, Paul Richardson, MD, of Dana-Farber Cancer Institute in Boston, Massachusetts, took the clinicians’ perspective and suggested that, at this point, “we’re not yet ready to apply it to everyday practice.”
“[P]atients who have a complete response (CR) and are MRD negative have longer progression-free survival (PFS),” Dr Landgren pointed out, “and there are indications that their overall survival (OS) is better than in those patients who are just CR and MRD positive.”
“My position on this is that MRD testing is absolutely ready for prime time in the research and regulatory arena,” Dr Richardson contended. “The question for me, as a clinician, in my clinic, is ‘Do I apply it to everyday practice?’ And I would simply suggest to you, at this point, we’re not ready for that.”
Yes—MRD is ready for prime time
Dr Landgren based his argument on 2 meta-analyses published in 2016 and 2017 that outline the importance of MRD status in newly diagnosed MM patients.
The first analysis (Landgren et al 2016) showed that MRD negativity was associated with better PFS (hazard ratio [HR]=0.35] and OS (HR=0.48) than MRD positivity.
“So using more simple language,” Dr Landgren said, “this means that MRD negativity reduces the risk of progression by 65%, and it also reduces the risk of dying by 52%.”
The second analysis (Munshi et al 2017) also associated MRD-negative status with superior survival outcomes for both PFS (HR=0.41) and OS (HR=0.57).
As further confirmation of the importance of MRD status, the International Myeloma Working Group last year published response definitions that include MRD negativity at a sensitivity of 1 in 105 cells or higher as the deepest level of treatment response in MM.
Dr Landgren drew on additional studies to support routine MRD testing in patient care.
The IFM Study Group found that, in newly diagnosed patients treated with lenalidomide, bortezomib, and dexamethasone followed by 1 year of lenalidomide maintenance, patients who received a subsequent transplant achieved superior outcomes compared to non-transplanted patients, in terms of CR (58% vs 46%) and 3-year PFS (61% vs 48%).
However, in patients who were MRD negative in both arms, the PFS rates were very similar, Dr Landgren said. And in terms of 3-year OS, there was no difference, at 88% in both arms.
The experience with daratumumab in relapsed/refractory patients exhibited a similar pattern.
The phase 3 POLLUX trial first showed that adding daratumumab to lenalidomide and dexamethasone was superior to lenalidomide and dexamethasone only, with a PFS at 18 months of 78% and 52%, respectively. This amounted to a 63% reduction in the risk of disease progression.
Investigators then took one more step forward, Dr Landgren said, and looked at MRD.
At a sensitivity of 10-5, almost 25% of patients on the 3-drug regimen were MRD negative, “which is kind of amazing,” Dr Landgren said. “This is a very big step forward.”
“If you break down the results by MRD status, which is not the primary endpoint of the study, you see very similar patterns for PFS for MRD negative patients in each of the 2 arms,” he continued.
This raises the question of whether attaining MRD negativity is more important than the treatment modality.
MRD negativity has implications for speeding drug approvals, developing more sensitive assays, and future treatment management, Dr Landgren said.
No—MRD is not ready for prime time
Dr Richardson acknowledged that MRD assessment is important. However, he pointed out a number of caveats regarding how MRD assessment would be applied in clinical practice to support his position.
“I’d simply suggest to you that, in day-to-day practice, the definition [of MRD] is somewhat fluid,” he said. “And it varies, obviously, between diseases and technology used.”
For most malignancies, Dr Richardson said, 109 to 1010 malignant cells are undetectable with conventional methods. These may or may not lead to a full clinical relapse within months or even years.
Using a sensitive technique to determine the presence of MRD could permit analysis of treatments that induce a greater depth of response or identify patients at risk of early relapse who need further treatment.
Dr Richardson enumerated hematologic malignancies that utilize MRD as secondary endpoints—acute lymphoblastic leukemia, acute myeloid leukemia, acute promyelocytic leukemia, chronic lymphocytic leukemia, follicular lymphoma, and mantle cell lymphoma.
In chronic myeloid leukemia, MRD is used as a primary endpoint that dictates practice.
“And I would applaud the field in that area because, obviously, molecular response accepted as an endpoint by FDA for second-generation TKIs has been a bedrock of that approval process, and it now applies in clinical practice,” Dr Richardson said.
“Obviously, that’s where we’d like to be, but I’d suggest to you, just again, with a certain amount of moderation and a certain amount of caution, that we may not be quite there yet.”
Dr Richardson suggested that MRD assessment in MM is less advanced than in leukemia and lymphoma.
“[W]e are currently at the point where MRD assessments are clearly secondary endpoints, an important research tool,” he said.
Some “remarkable combination therapies,” he added, have abrogated some of the “extraordinary genetic complexity” in MM.
“The critical point here, though, is that, while we’re more successful in terms of these triplets and quadruplets and now with the introduction of monoclonal antibodies and similar approaches, we’re able to throw a bigger net around the disease,” Dr Richardson said.
“We’re not able to eradicate it completely, and cure remains, in myeloma, frankly, evasive. And I think that’s a critical point.”
Dr Richardson reviewed various strategies for molecular response monitoring, from flow cytometry to polymerase chain reaction and next-generation sequencing, noting that there is variance in applicability and sensitivity.
For example, the limits of detection among 91 labs ranged from 0.10% to 0.001%.
Dr Richardson returned to the “very robust” meta-analysis by Munshi and colleagues discussed by Dr Landgren.
While the authors’ analysis demonstrated that MRD is predictive of both longer PFS and OS, they concluded that the evidence supported MRD as an endpoint and research tool in clinical trials.
“So I would humbly suggest perhaps it’s not ready for clinical prime time yet,” Dr Richardson said.
He also referred to the IFM Study Group trial described by Dr Landgren, calling it a “critical forward effort.”
“[W]hat’s so interesting is that there was no difference in overall survival,” Dr Richardson said. “Now, that’s a very important point as we soberly look at these data and judge what they mean for each patient.”
And so Dr Richardson stood by his assessment that MRD is not yet a standard of care but may be one day.