Measurable residual disease in hematologic malignancies: a biomarker in search of a standard

Over the past decade, measurable residual disease (MRD) has emerged as a critical tool for detecting and monitoring a variety of cancers, but particularly hematologic malignancies. The rapid adaptation of this novel approach to monitoring disease status is intuitively appealing: it offers significantly greater sensitivity than traditional methods for detecting the small population of malignant cells that persist after treatment, and which are undetectable by standard approaches such as monitoring abnormalities on radiographic scans or assessing morphologic or karyotypic changes from bone marrow sampling. Overall, MRD reflects the cumulative effect of tumor biology, treatment tolerability and safety. Commonly used modalities for detecting MRD include multiparametric flow cytometry (MFC), next-generation sequencing (NGS), and polymerase chain reaction-based methods (PCR). While complete remission-variably defined depending on the cancer-has traditionally been an immediate treatment goal, MRD analysis has the potential of refining this concept by detecting minimal residual disease that conventional assessments may miss, calling into question the true "completeness" of the remission. While MRD negativity does not equate to cure-as some patients still relapse-it represents a meaningful advance over traditional remission criteria, providing deeper insights into treatment response and relapse risk while we continue refining its predictive power. As often occurs in science, technology has developed more quickly than the confidence in how to apply and react to it clinically. Thus, the role of MRD in clinical management is still being defined and varies by disease. Broadly speaking, what is clear is that MRD status is nearly universally correlated with prognosis and risk stratification across hematologic malignancies: MRD positivity signifies residual disease and is associated with worse outcomes, whereas MRD negativity suggests a low or undetectable (though not necessarily absent) disease level and a better prognosis. Additionally, earlier disease clearance tends to portend improved outcomes. This has face validity and aligns with intuition-if the disease is sensitive to treatment, it is likely to be eradicated quickly and completely, making it undetectable. In some diseases, the utility of MRD extends beyond prognosis. MRD can be used to evaluate treatment response, and can guide therapy personalization-including escalation, de-escalation, or optimization of therapy duration. Further, it can be used to monitor disease in both pre- and post-transplant settings, enabling earlier relapse detection before clinical or routine laboratory signs of disease appear. As MRD technologies advance, their role in clinical trials is also expanding, serving as a surrogate endpoint for more clinically meaningful outcomes such as survival in some malignancies, and as a biomarker in early-phase drug development. As we incorporate MRD assessments into routine clinical management, though, questions remain: For disease that isn't readily eliminated, should MRD negativity always be a goal? Across diseases, does intervention at an MRD-positive state improve survival more than waiting for morphologic or radiographic relapse? And critically, will reacting to an MRD positive state in some diseases lead to intensified therapy and overtreatment despite a low risk of progression?