FDA Draft Guidance Proposes MRD as Accelerated Approval End Point in Multiple Myeloma

The FDA released draft guidance in January proposing minimal residual disease (MRD)-negative complete response as a primary end point for accelerated approval of drugs for multiple myeloma. The guidance defines MRD negativity as fewer than 1 myeloma cell per 1 million bone marrow cells, assessed by flow cytometry or sequencing following a complete response.

Under the framework, MRD functions as an intermediate clinical end point—reasonably likely to predict clinical benefit—rather than a validated surrogate. Consequently, the guidance is scoped narrowly to regulatory decision-making for accelerated approval and does not address individual treatment decisions or use in the maintenance setting.

The guidance, developed collaboratively across the Oncology Center of Excellence, Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research, and Center for Devices and Radiological Health, outlines strict requirements for trial design. Randomized trials are preferred over single-arm studies because they allow for the continued investigation of progression-free survival and overall survival as secondary or safety end points. Requirements include robust data collection, predefined statistical analyses, standardized bone marrow aspirate assessments to minimize bias, and complete trial enrollment before the targeted end point is analyzed for approval. Although these recommendations are not currently legally binding, they reflect the agency's evolving thinking on quantifying treatment depth.

Nicholas Richardson, DO, MPH, of Precision for Medicine, discussed the implications of the draft guidance. He identified calibration failures as one of the most persistent challenges, stressing that assays must be analytically valid and that sponsors should proactively troubleshoot known risks. The goal is to generate MRD data that are clinically meaningful on a patient-by-patient level and robust enough for inclusion in an FDA label.

Richardson also described the single-trial model as a potentially powerful tool when used thoughtfully. Rather than conducting a single-arm trial in a heavily pretreated population for accelerated approval, followed by a separate randomized trial to confirm benefit, a single randomized trial in an earlier treatment line, such as patients with 1 to 3 prior lines, could serve both purposes. An early MRD-negative complete response assessment between arms could support accelerated approval, while long-term follow-up would confirm progression-free survival and overall survival. He emphasized that sponsors must fully power this MRD assessment, prespecify a rigorous statistical analysis plan, ensure complete enrollment before the primary MRD readout, and engage proactively with the FDA throughout.

On the question of a timeline for finalizing the guidance, Richardson said it is difficult to predict. The FDA reviews every public comment submitted, and the depth and substance of those responses will shape the path to a final document. He encourages all stakeholders—drug developers, patient advocates, and patients alike—to engage with the draft guidance and submit comments, noting that this input will produce the strongest possible guidance. Public comments on the draft are being accepted through March 23, 2026.