Don’t let the genie out of the bottle!

In this issue of Blood, 2 companion articles from Guerrero et al¹ and D’Agostino et al² explore predictors of disease resurgence despite obtaining measurable residual disease negativity (MRD_neg) with first-line therapy in newly diagnosed multiple myeloma (MM).

It is amply evident that achieving an MRD_neg status is the most impactful outcome from the treatment of MM, as it is for many hematological cancers, improving progression-free survival (PFS) and, importantly, overall survival (OS).³ Although MRD_neg does not translate into a cure in MM, it is likely that the profound reduction in the clone size puts a brake on the clonal evolution, leading to improved OS.⁴ Notably, the therapeutics have kept pace with the increasing sensitivity of MRD detection technology, with the current treatments leading to an unprecedented MRD_neg rate even when defined stringently using a cutoff of 10⁻⁶. In the absence of MRD_neg not translating to a cure, it is essential to demonstrate the value of maintaining MRD_neg status for considerable periods of time, the concept of sustained MRD_neg, which is already part of the International Myeloma Working Group response criteria.⁵ In fact, many studies have examined the value of sustained MRD_neg, typically using 6- and 12-month intervals, and demonstrated the positive impact on PFS and OS.⁶ What has become disappointingly obvious, with longer follow-up of recent studies, is that nearly all patients eventually relapse, with a substantial proportion of those with MRD_neg sustained for a year or more losing that status as a prelude to eventual biochemical progression and or reappearance of hypercalcemia, renal failure, anemia, and bone disease (CRAB) features. So, what are the drivers of this early failure or inability to sustain the success achieved in the form of MRD_neg? Identifying these factors is essential, as this allows us to tailor the intervention for a given patient. These can only be studied in the context of well-designed trials using consistent time points and methodology for MRD testing, not easily achieved in routine clinical practice.

In this context, the results of 2 different analyses published in this issue of Blood shed light on this important clinical problem. D’Agostino et al studied patients enrolled in the FORTE trial, which used carfilzomib and dexamethasone with either lenalidomide or cyclophosphamide for induction and posttransplant consolidation, followed by lenalidomide with or without carfilzomib for maintenance (transplant was used in two-thirds of patients).⁷ They systematically examined the MRD status every 6 months, starting with the initiation of maintenance to understand the kinetics of MRD_neg. In their analysis, the presence at baseline of high-risk cytogenetic abnormalities and circulating tumor cells and a longer time to achieve MRD_neg were associated with a higher risk of MRD recurrence after having sustained it for a year. Guerrero et al did a similar analysis using a Spanish trial using a bortezomib, lenalidomide, and dexamethasone regimen for induction and posttransplant consolidation, followed by lenalidomide with or without ixazomib for maintenance. MRD testing was also done at prespecified time points in this study, albeit at longer intervals. With a longer follow-up, the second study identified International Staging System stage 3 and the presence of circulating tumor cells from baseline, and a longer time to achieving MRD_neg, as predictors of unsustained MRD_neg, while high-risk cytogenetics approached significance. It is remarkable that the 2 analyses came to similar conclusions, highlighting the overarching impact of disease biology (high-risk cytogenetics, circulating cells) and depth of response. The impact of the time to MRD_neg is an important finding, opening the potential for early intervention strategies.

What messages do these studies have for our understanding of disease biology and the questions that need to be answered in future studies? Clearly, MM is a heterogeneous disease, with disease biology being the primary determinant of differential outcomes, and despite improvements in therapeutics, those with high-risk diseases continue to do poorly.⁸ Achievement of deep responses do have a counterbalancing effect, but do we need an even deeper response for these patients, potentially MRD_neg measured at 10⁻⁶ level? Can we achieve this by increasing the intensity of therapy, or do we have to strategically incorporate novel therapies with mechanisms of action targeted to the altered biology? Identifying the predictors of early failure is an essential first step toward the design of trials exploring these options. Both these trials explored the concept of limited-duration therapy, and it is likely that continued treatment, especially with 2-drug combinations, may have reduced the risk of, or at the least prolonged the time to, MRD recurrence. In fact, there was a decreased risk of resurgence in the FORTE trial while patients were on 2-drug maintenance. This brings us to the question of response-adapted treatment strategies in MM, an area of active investigation. Considering the disease risk and the depth of response together in a composite manner will allow us to tailor the intensity and duration of therapy for the individual patient, permitting treatment-free intervals for those at the lowest risk, a concept that will require validation in trials. Interestingly, the level of MRD at resurgence was predictive of subsequent outcomes, likely a reflection of the proliferation rate of the clone.

An important observation is that in many patients, disease progression happened prior to MRD resurgence being detected, despite the frequent measurements of MRD status, highlighting the inability of less frequent marrow-based assessments to accurately predict impending relapse. This finding has implications as we explore the concept of early intervention prior to the development of CRAB features or frank biochemical progression, testing the hypothesis that earlier resumption of treatment may lead to improved survival. It emphasizes the importance of less invasive, blood-based assessments for early detection of the disease resurgence, made feasible with the use of sensitive mass spectrometric approaches.⁹ Some of these patients presented with extramedullary disease, highlighting the need for periodic imaging during monitoring.¹⁰ Nevertheless, the period between MRD resurgence detected by any method and disease progression prompting treatment currently offers opportunities for testing early intervention strategies.

It is important to note that a large proportion of those who lost the MRD_neg did so despite having sustained it for over a year. What is unclear is whether sustenance of MRD_neg above a certain period will eliminate the risk completely, a surrogate for a cure, or if recurrence is inevitable. Longer follow-up from studies like these will help answer this critical question.

Conflict-of-interest disclosure: S.K. reports consulting/advisory board participation (with no personal payments) with AbbVie, BMS, Janssen, Roche-Genentech, Takeda, Pfizer, Loxo Oncology, K36, Sanofi, ArcellX, and Beigene.