Next-Generation Sequencing MRD Detection Predicts Outcomes in AML Patients Undergoing Allogeneic Stem Cell Transplantation

Overview

Acute myeloid leukemia (AML) is an aggressive blood cancer that often requires intensive treatment followed by allogeneic hematopoietic stem cell transplantation (allo-SCT), also called an allogeneic stem cell transplant. Even when patients appear to be in remission before transplant, tiny numbers of leukemia cells can remain in the body. These cells are known as measurable residual disease (MRD), and they can strongly influence the risk of relapse and survival after transplantation.

A new study published in Haematologica evaluated a cost-conscious next-generation sequencing (NGS) approach for MRD detection in AML. The results suggest that this method can identify patients at higher risk of poor outcomes before transplant, potentially helping doctors personalize treatment more effectively.

Why measurable residual disease matters in AML

In AML, achieving remission after chemotherapy does not always mean that all leukemia cells have been eliminated. Standard tests may miss very small amounts of disease. MRD testing is designed to detect these hidden cells and estimate relapse risk.

This information is especially important before allo-SCT, because transplant is a major treatment with both curative potential and significant risks. If MRD is still present before transplant, the chance of the leukemia coming back afterward is usually higher. On the other hand, patients with no detectable MRD often have better outcomes.

Traditionally, MRD in AML has been assessed using flow cytometry or PCR-based methods, but these approaches are limited to certain patients or mutations. NGS can examine many genes at once, making it broadly useful, but standard NGS methods can be expensive and difficult to implement routinely in some settings.

What the researchers studied

The investigators developed a targeted NGS MRD test using single-molecule molecular inversion probes, or smMIPs. This technology allowed them to sequence 92 genomic regions across 33 genes commonly involved in AML.

They applied the assay to 93 patients with AML who were in remission before allo-SCT. The main goal was to determine whether MRD detected by this method could predict outcomes after transplantation.

For this study, MRD positivity was defined as the presence of at least one non-DTA mutation with a variant allele frequency, or VAF, of 0.5% or higher. Non-DTA refers to mutations outside DNMT3A, TET2, and ASXL1, because mutations in these genes can sometimes reflect age-related clonal hematopoiesis rather than active leukemia.

Main findings

The study found that MRD positivity before transplant was associated with significantly shorter overall survival after allo-SCT. In other words, patients whose blood or marrow still showed residual leukemia-related mutations had worse outcomes.

In statistical analysis, NGS-based MRD remained an independent predictor of poor overall survival even after accounting for other clinical factors. The reported hazard ratio was 4.58, indicating a substantially higher risk of death among MRD-positive patients. The association was highly significant, with p = 0.002.

The researchers also found that the intensity of conditioning, meaning how strong the pre-transplant chemotherapy or radiation regimen was, did not clearly change outcomes in MRD-positive patients in this retrospective cohort. This suggests that simply using a more intensive conditioning regimen may not be enough to overcome the risk associated with residual disease, although larger prospective studies are needed to confirm this.

Mutation patterns before and after transplant

The study also compared mutations found at diagnosis with those detected before transplantation. Concordance varied by gene, meaning that some mutations persisted while others disappeared or changed in detectability over time. Overall, variant allele frequencies were consistently lower before transplant than at diagnosis, as expected after treatment.

In three patients, multiple low-VAF clustered variants in RUNX1 and TET2 were found before transplant. The pattern raised the possibility of treatment-induced mutagenesis, meaning that therapy may have contributed to the appearance of new low-level genetic changes. While intriguing, this finding should be interpreted cautiously and requires further investigation.

Why this assay is important

One of the most notable aspects of this study is the practical design of the test. The smMIP-based assay was reported to have low library preparation costs of about 8 euros per sample, making it much more affordable than many conventional NGS workflows.

That affordability matters because MRD-guided care could help more patients receive the right treatment at the right time. For example, MRD-positive patients might benefit from closer monitoring, post-transplant maintenance therapy, donor lymphocyte infusions, clinical trials, or other MRD-directed interventions depending on the clinical context.

The broad applicability of the assay is another strength. Because it covers many AML driver genes, it may be useful across a wide range of patients rather than only those with a single well-defined mutation.

Clinical implications

This study supports the growing role of molecular MRD testing in AML, particularly before allo-SCT. If validated in larger cohorts, this approach could help clinicians better identify patients at high risk for relapse and tailor management accordingly.

Potential uses include:

– refining transplant risk stratification;
– selecting patients for intensified post-transplant surveillance;
– identifying candidates for maintenance therapy;
– supporting enrollment in MRD-directed clinical trials;
– improving allocation of limited healthcare resources in settings where cost is a major barrier.

Importantly, MRD testing should not be used in isolation. Decisions about transplantation and post-transplant care still depend on age, comorbidities, disease subtype, cytogenetics, prior therapies, donor availability, and patient preferences.

Limitations

As with any retrospective study, there are limitations. The number of patients was relatively small, and treatment decisions were not assigned randomly. That means unmeasured factors may have influenced outcomes.

In addition, the study focused on a defined panel of AML-related genes. Although broad, this approach may still miss some relevant abnormalities. More research is also needed to determine how best to integrate NGS MRD results with flow cytometry, PCR, and other clinical risk factors.

Finally, the definition of MRD positivity can vary across studies and platforms. Standardization will be essential before NGS-based MRD can be used consistently across centers.

Bottom line

This study shows that a low-cost, smMIP-based NGS assay can detect clinically meaningful measurable residual disease in AML before allogeneic stem cell transplantation. Patients with detectable MRD had worse overall survival, and this effect remained significant even after adjustment for other factors.

The findings point toward a future in which more accessible molecular testing helps guide transplant decisions and post-transplant care in AML. With further validation, this type of assay could expand MRD-guided treatment to more patients, including those in resource-limited settings.