Optimizing usage of measurable residual disease (MRD) for treatment decision making in acute myeloid leukemia

Acute myeloid leukemia (AML) is a malignancy affecting bone marrow, characterized by abnormal cell maturation and proliferation. Initial treatment involves intensive chemotherapy aiming for complete remission (CR), followed by consolidation therapy. Consolidation options include chemotherapy alone, autologous stem cell transplant (auto-SCT), or allogeneic stem cell transplant (allo-SCT). Selecting the appropriate consolidation therapy balances anti-leukemic efficacy with safety concerns. While allo-SCT reduces relapse risk, it also carries significant morbidity, prompting cautious use. Measurable residual disease (MRD) is increasingly influential in AML management, indicating higher relapse risk when detected pre-consolidation. Measuring MRD is subject to strict protocols and guidelines to ensure consistent and accurate identification and detection of MRD. In addition, it is important that results are harmonized between centers so that results can be compared. Within Europe, this harmonization process is led by the European LeukemiaNet (ELN). Chapter 2 discusses the technical aspects of flow MRD, which is a consensus on the entire process from administration to implementation of MRD. In addition, when MRD is used for clinical decision-making, it must comply with reliability standards and rules set by the European Union, called in vitro diagnostics rules (IVDR). In Chapter 3 we show that our MRD assay meets all requirements set by the IVDR and therefore qualifies for clinical use. While using MRD to guide treatment remains complex, following these guidelines already provides a solid basis for use in practice. In the HOVON/SAKK-132 study in Chapter 4, the choice for SCT is based on the MRD result after initial chemotherapy. With an MRD-positive result in AML patients with an intermediate-risk, there is an increased chance of the disease relapsing and an allo-SCT is advised. Whereas, when a patient is classified as MRD-negative, de-escalation is done by omitting an allo-SCT and switching to an auto-SCT. In this detailed analysis of the clinical study, we showed that there was no difference between the two groups in the risk of relapse and diminished survival, despite the fact that the MRD-negative group had received a less severe consolidation therapy. Therefore, we concluded that MRD-negative patients could safely receive a non-allogeneic treatment, challenging the conventional belief that all intermediate-risk patients should undergo allo-SCT in CR1. In Chapter 5, we showed that early MRD assessment after one chemotherapy course shows prognostic value, aiding risk stratification and transplant planning. In addition, a sequential approach of two time points can increase the sensitivity of MRD assessment. Primitive cell-based MRD assessment (PM-MRD) may identify high-risk patients missed by standard MRD methods, as showed in Chapter 6. In Chapter 7 we investigated the effect of hemodilution on MRD and the value of the different formulas. The MRD rate decreased statistically significantly after aspirating the first milliliters of bone marrow, underlining the importance of recognizing hemodilution. The mast cell population can be used to discriminate between bone marrow and peripheral blood. In Chapter 8 we investigated whether autologous stem cell apheresis products (ASCAP) can serve as an alternative source for measuring MRD instead of bone marrow. ASCAP samples offer an alternative to bone marrow for MRD assessment, correlating with relapse risk. However, sensitivity remains comparable, suggesting cautious interpretation. In the future, the continuous refinement of measurement of MRD and well-designed studies with MRD-driven clinical decision-making should contribute to the development of MRD from a prognostic marker to a predictive marker for the individual patient, as discussed in Chapter 9. Until then, MRD can be a valuable tool in specific scenarios, for example in the selection of suitable patients for withholding an allo-SCT. Nevertheless, such decisions should be made carefully, taking into account the broader clinical context of the individual patient.