The FDA recently released a draft guidance to assist with the use of minimal residual disease as a biomarker of tumor burden quantification as well as to clarify issues pertaining to this use of MRD in regulatory submissions, including in clinical trials, or to support marketing approval of drugs and biologics for treatment of specific hematologic malignancies. In this discussion, experts say this represents a timely and significant move toward advancing the role of MRD as a regulatory endpoint.
Kenneth C. Anderson, MD
The FDA recently released a draft guidance to assist with the use of minimal residual disease (MRD) as a biomarker of tumor burden quantification.1The guidance seeks to clarify issues pertaining to this use of MRD in regulatory submissions, including in clinical trials, or to support marketing approval of drugs and biologics for treatment of specific hematologic malignancies.
“This is useful guidance and will both identify contexts where MRD can be used and where we need more data for it to be extended more broadly,” said Kenneth C. Anderson, MD, Kraft Family Professor of Medicine at Harvard Medical School and program director of the Jerome Lipper Multiple Myeloma Center and LeBow Institute for Myeloma Therapeutics at Dana-Farber Cancer Institute in Boston, Massachusetts. Anderson has been actively engaged with the FDA and the Foundation for the National Institutes of Health, spearheading the MRD concept, especially in multiple myeloma (MM).
Other experts agree that this represents a timely and significant move toward advancing the role of MRD as a regulatory endpoint. “The guidance document is an important step going forward,” said C. Ola Landgren, MD, PhD, chief of Myeloma Service at Memorial Sloan Kettering Cancer Center in New York, New York. “It is a strong signal that the regulatory agency, academia, and pharma are aligned in the interpretation of the importance of MRD for hematologic malignancies.”
The guidance document “demonstrates that the FDA recognizes the need for us to move beyond conventional endpoints,” said Shaji Kumar, MD, a consultant and professor of medicine in the Division of Hematology, Department of Internal Medicine, at Mayo Clinic in Rochester, Minnesota. “[It is] quite detailed, especially bringing out the potential differences between disease statesand more importantly, from the MM perspective, they have nicely outlined some of the key questions the field has been grappling with in terms of ongoing efforts to prove the surrogacy of MRD.”
C. Ola Landgren, MD, PhD
Reduction of tumor burden is a traditional and essential indicator of treatment response. MRD can serve as a surrogate marker of tumor burden and may indicate poor tumor reduction or regrowth of highly proliferative cancer cell clones following treatment. The assessment of MRD, a measure of the depth of the response, recently emerged as a potent concept in detecting and defining submicroscopic disease, propelled by advances in cellular and molecular technology platforms and complete responses (CRs) with newer therapies in a larger proportion of patients. Importantly, MRD detection correlates significantly with clinical outcomes in many hematologic malignancies, further substantiating the need for additional data regarding the appropriate method for MRD assessment and identification, standardization of ideal disease-specific biomarker thresholds indicative of MRD, and expansion of assessments in clinical studies and practice.2-12
The association between MRD negativity and clinical outcomes, including progression-free survival (PFS) and overall survival (OS), has been described in many hematologic malignancies, including MM, acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), follicular lymphoma, chronic lymphocytic leukemia (CLL), mantle cell lymphoma (MCL), and chronic myeloid leukemia (CML).2-10In MM, for example, a recent meta-analysis showed that patients who achieved MRD negativity had better PFS (HR, 0.41) and OS (HR, 0.57) compared with MRD-positive patients. This association was reiterated in patients who achieved CR; MRD-negative status in patients who achieved CR was associated with significantly improved PFS (HR, 0.44) and OS (HR, 0.47).3In MCL, an analysis of a subset of patients from the European MCL Younger and MCL Elderly studies showed superior PFS in patients with undetectable MRD compared with patients who developed MRD-positive status within 6 months.11MRD response at the end of induction and at the end of the consolidation treatment or an equivalent phase have been established as key prognostic indicators in B-cell ALL (B-ALL) and T-cell ALL.6In a study of adult patients with B-ALL, MRD-negative statusassessed using multiparameter flow cytometry (MFC)—at CR was associated with improved disease-free survival and OS.12
The FDA noted 2 major mechanisms that are relevant to the regulatory acceptance of MRD as a surrogate endpoint: the formal Drug Development Tool qualification process and a collaborative review process. With either approach, establishment of MRD as a surrogate endpoint requires evidence of underlying biological rationale for the association, epidemiological data supporting the MRDclinical outcome association, and correlation between treatment-mediated effects on MRD status and treatment response in clinical studies. The FDA also listed the statistical criteria for validating MRD surrogacy in meta-analyses of data from clinical trials.1
MRD may serve as a diagnostic, prognostic, predictive, efficacy-response, or monitoring biomarker. Several factors may impinge on how MRD status is used in a particular malignancy/patient/treatment context; for instance, MRD status can be assessed prior to patient selection for a given treatment, where it serves as a biomarker predictive of response.1
The recent expansion of the FDA approval of blinatumomab (Blincyto) for the treatment of adults and children with B-cell precursor ALL who are MRD positive exemplifies the use of MRD status to select patients and monitor treatment response.13,14The approval expansion was based on data from the BLAST trial (NCT01207388), which showed that 78% of adult patients with MRD-positive ALL in hematologic remission after chemotherapy achieved a complete MRD response with blinatumomab and that the MRD response correlated with relapse-free survival and OS.15 The primary endpoint was complete MRD response status after 1 cycle of blinatumomab.15The subsequent FDA approval expansion was the first instance of the use of MRD as a biomarker in a regulatory submission.13
Shaji Kumar, MD
More recently, the FDA-approved expansion of the labeling information for venetoclax (Venclexta) with the addition of MRD-negativity data from the MURANO trial (NCT02005471) was also the basis for the approval of venetoclax and rituximab as the first chemotherapy-free combination for patients with previously treated CLL.16,17The data showed that among patients with relapsed or refractory CLL, the venetoclax-rituximab combination yielded a significantly higher 2-year PFS rate than the bendamustine-rituximab combination (HR for progression or death, 0.17).17MRD status in CLL in this context could be characterized as a treatment response biomarker.
Across hematologic malignancies, MRD positivity has been correlated with relapse and disease progression.2-10To date, MRD status has been added to the response criteria in MM in patients who have achieved CR,18and this may serve as a paradigm for adding MRD assessment as a tool to evaluate response in other hematologic malignancies. Kumar, who coauthored the most recent International Melanoma Working Group (IMWG) consensus criteria for response and MRD assessment in MM, noted, “For MM, the field has espoused a forward-thinking approach in revising the response criteria. We needed to include the MRD assessment to ensure a uniform approach…before we can move forward with the ways in which MRD assessment can and should be utilized as a platform in clinical practice.”
The current considerations in the FDA draft guidance for sponsors of clinical studies considering MRD assessment as a biomarker or an endpoint are summarized inTABLE 1. It is important to note that the considerations are a starting point, and sponsors are directed to submit justifications for additions or alterations. As additional data on MRD emerge, the details of the aspects described may be revised. An example would be restriction of MRD assessment in patients with CR alone in MM.1“The guidance recommends measurement of MRD only in the setting of CR,” Anderson said. “[Although] this is where most data are obtained now, there are examples of where MRD can be negative even in patients with less than CR, and many ongoing studies are evaluating MRD even in these patients. These ongoing studies will help define the value of measuring MRD more broadly.”
Landgren, who was also a coauthor on the IMWG consensus response criteria and cochairs the Advances in MRD Testing in Myeloma Meeting, elaborated: “The recommendations of experts in the MRD/MM field at the sixth annual meeting were that MRD testing should be conducted both in patients with CR and very good partial response [VGPR]. If we look at the recently published data, there is information regarding this that the field has not been able to fully digest yet.”
He pointed to studies showing that when treating MM with chimeric antigen receptor (CAR) T-cell therapy, most patients were MRD negative 1 month after receiving the therapy. However, nearly all patients also had detectable myeloma protein (MP), in some instances at levels unaltered by the treatment. “But importantly,” Landgren said, “when the protein was tested over subsequent months, most of the patients who were MRD negative at 1 month, although they had high initial levels of MP, had progressive lowering of MP levels, and in those with CR, the levels became undetectable in 6 to 12 months.”
Landgren interprets those data to indicate renal clearance of MP, which from a biomarker standpoint represents a false-positive signal for tumor burden. He added that these data may be a function of the rapid disease clearance with newer therapies. The expert consensus is that MRD testing may be applied in patients with CR and VGPR who were treated with combination therapies or in cases of rapid clearance, such as with CAR T-cell therapy, and higher MP thresholds for MRD testing may be appropriate. “This is an evolving area in MM translational medicine,” he added.
MRD assessment methodologies are disease specific, even though the common goal is to estimate the presence and frequency of malignant cells. Traditional methods for detecting malignant cells include detection of hematologic malignancyspecific cell surface markers, using flow cytometry–based methods or cancer cell–specific intracellular markers, such as RNA transcripts detected using reverse transcription-polymerase chain reaction (RT-PCR).18-25The FDA acknowledges being agnostic to the MRD assay technology platform, focusing instead on the prespecification of the platform details, including the procedure, reagents, and analysis, as well as sensitivity. Four general technologies have been identified for MRD assessment in hematologic malignancies: MFC, next-generation sequencing, quantitative RT-PCR of specific gene fusions, and allele-specific oligonucleotide polymerase chain reaction.1In general, MRD assessment methods have a sensitivity in the range of 10−4to 10−6. The ideal method for MRD assessment in every patient subpopulation, treatment, and disease state is one of the biggest challenges going forward before MRD assessment can be standardized and applied uniformly in clinical studies and practice. The key attributes and considerations of current and emerging technologies for MRD assessment are summarized inTABLE 2.25-29These more sensitive technologies include deep sequencing of circulating cell-free DNA, droplet digital polymerase chain reaction, genomic analysis of circulating tumor cells, mass spectrometrybased methods, and functional imaging methods such as immune–PET and MRI.2,18-25Landgren noted that his group translates the use of immune-PET to monitor residual disease from mouse models to humans. Additional studies are needed to clarify the utility and feasibility of immune-PET and other methods in clinical practice.
The sensitivity and specificity of the assay for the marker that is being analyzed as an indicator of the presence of malignant cells, as well as the sampling site, protocol, and timing, need to be established for each disease state and circumstance to ensure that MRD assessments can be applied uniformly and the data across trials can be compared.1Despite its association with clinical outcomes, including MRD assessment in clinical management of hematologic malignancies remains a major challenge, due to differences in the assay attributes and preferred analytical methods employed between routine laboratories. MRD assessments and their role in management of hematologic malignancies, including AML, CLL, MM, and lymphomas, are being investigated vigorously and extensively.12,25-30The FDA draft has provided a starting framework for MRD assessment methodologies, including recommendations for key aspects to consider when including these assessments in clinical trials and supporting data for new or expanded approvals, all of which will help clarify how, when, where, and in whom MRD testing can and should be applied in clinical practice.1
MRD testing is here to stay; the potent concept may be extrapolated beyond hematologic malignancies. Speaking to the future of MRD assessments on the heels of the key draft guidance from the FDA, Landgren said, “I am very excited; I think that there will be 2 major deliverables in the near future. One will be that MRD will become a regulatory endpoint for drug approval. The consequence of that is that drugs can progress to market more quickly, which will be great for patients, providing faster access to additional and improved therapeutic options. The second deliverable will be that with additional studies on the utility of MRD-guided treatment, what we might end up with is patients who either achieve MRD negativity with fewer cycles or lower intensity of treatment, and you can expose patients to less toxicity.”
Kumar said, “I think MRD is going to be a very useful tool for us to adapt the treatment intensity and duration, an aspect that needs to be demonstrated in prospective clinical trials. We are already designing trials to address the role of MRD status in guiding treatment decisions in MM.” Two imminent trials led by the Eastern Cooperative Oncology Group and SWOG, formerly the Southwestern Oncology Group, are directly addressing the role of MRD status in guiding treatment, he said.
Anderson stated, “I do think that this guidance is reflective of the current state of the science and practice and will help fast-forward new drug development.