Minimal Residual Disease (MRD) Data in Hematologic Malignancy Drug Applications and Prescribing Information: FDA Analysis

Introduction: In the past decade, multiple studies have reported the prognostic and predictive value of MRD status in specific hematologic malignancies (HM). Because clinical trials are increasingly incorporating MRD status as a biomarker and efficacy endpoint, the adequacy of the MRD data to inform the prescribing information (PI) is relevant for the design and conduct of pivotal clinical trials. We present an analysis of the trends in inclusion of MRD data in pivotal trials in HMs and regulatory decisions made by the U.S. Food and Drug Administration (FDA).

Methods: We reviewed FDA internal databases for original and supplemental new drug applications (NDAs) and biologics licensing applications (BLAs) submitted 1/2014-12/2020 to support approval of therapies (drugs, biologics, and cellular therapies) for HM. MRD data were evaluated for two time periods to inform potential trends: 1/2014-6/2017 (period 1) and 7/2017-12/2020 (period 2). Clinical study reports, selected datasets, FDA clinical reviews, and the proposed and approved PIs were examined for inclusion of MRD data, and FDA assessments of the adequacy of the MRD data for inclusion in the PI were reviewed.

Results: Of 196 NDAs or BLAs involving HM submitted between 2014-2021, 53 (27%) had MRD data, including 53 pivotal trials. The trials included patients with chronic lymphocytic leukemia, chronic myeloid leukemia, acute myeloid leukemia, acute lymphocytic leukemia, and multiple myeloma. Twenty-one applications and pivotal trials with MRD data were submitted in period 1, and 32 applications and 35 trials were submitted in period 2. Three trials were resubmitted in period 2. MRD evaluation was specified as a secondary and exploratory endpoint in 35 (66%) and 19 (36%) of the trials, respectively. Of the 53 trials, MRD data was proposed by the Applicant for inclusion in the PI in 41 (77%) but was ultimately included in 25 (47%). Of the trials for which MRD data was proposed in labeling, MRD data were deemed adequate by FDA in 81% of studies in period 1 (13/16) and 48% of studies in period 2 (12/25). MRD assays in the PI included polymerase chain reaction, flow cytometry, and next-generation sequencing in 18 (72%), 5 (20%) and 4 (16%) of the trials, respectively, with the clinical threshold for test positivity ranging from 10 ^-3 to 10 ^-5. For 11 trials with MRD data in the PI (44%), the MRD was evaluated regardless of clinical response, and in 14 trials (56%) MRD was evaluated in patients achieving a specific clinical response. The leading reasons for excluding MRD data from the PI were analytical and test validation deficiencies (e.g., incomplete test characteristics data, lack of test validation overall or in that disease) followed by performance issues (e.g., high amount of test failure, inability to identify a clone) and issues with trial conduct or design (e.g., inadequate data collection, statistical issues).

Conclusion: A quarter of HM drug applications, including 53 pivotal trials, submitted to the FDA between 2014-2020 included MRD data. Characterization of regulatory actions showed that despite the increasing number of submissions proposing MRD data for inclusion in the PI, rates of inclusion of MRD data in the PI did not reflect this increase. Improvements in assay validation and performance characteristics, robust collection of MRD data, and appropriate statistical planning can enable greater representation of MRD data in prescription drug labeling.