Artificial intelligence may be leveraged to improve minimal residual disease measurements for patients.
A new review outlines the role of minimal residual disease (MRD) assessments via blood component sampling by liquid biopsies (LBs) in patients with myeloid malignancies.
Blood components that undergo molecular analysis “can be used as a powerful tool for prognostic and predictive purposes in myeloid malignancies,” researchers explained in Frontiers in Oncology.1
In addition, evolving evidence about the quantification and identification of cell- and gene-based biomarkers for myeloid malignancies may help monitor treatment response.
Although MRD monitoring via LBs is not standard in myelodysplastic syndrome (MDS), the area is being actively investigated, and in the future, these biopsies could replace more invasive bone marrow biopsies.
LBs have the benefit of being noninvasive and the ability of yielding a host of information, including the burden of the disease and the tumor’s heterogeneity and evolution.
MRD, which refers to the persistence of malignant cells in a patient who underwent a seemingly successful first treatment, is also beneficial in the prognostication, disease monitoring, detection of early relapse, guiding of postremission treatment, and measurement of treatment efficacy for myeloid malignancies. Molecular analysis can be carried out via flow cytometry or sequencing techniques. For flow cytometry, fluorescently labeled antibodies are used to analyze the protein expression on the cell surface or in the cytoplasm. This can help determine the immunophenotype of cell lineages and is suitable for MRD detection in acute myeloid leukemia (AML) and MDS, the authors wrote.
Alongside advancements in LBs, next-generation sequencing (NGS) is being used to monitor posttherapy responses in patients with myeloid malignancies, including AML.
High-sensitivity flow cytometry is the most common technique to detect circulating tumor cells in AML. However, NGS or quantitative reverse transcriptase polymerase chain reaction can also be used to identify a targetable mutation for surveillance of early relapse in these patients.
The quantification of circulating tumor cells (CTCs), which serves as a method of detection in LB, has also been incorporated into the MRD AML protocol.
But when it comes to the clinical application of these biomarkers, concerns remain due to the need for more standardization and lack of studies investigating the biomarkers’ specificity, the researchers said.
MDS holds the potential to advance to AML, and “there have been recent developments in the use of LBs for MRD monitoring despite several obstacles relating to the standardization of such techniques.”
For patients with BCR-/ABL-negative myeloproliferative neoplasms, a rare subtype of the disease with no known treatment,2 LB remains an experimental research tool.
Overall, the applicability of MRD using LB is limited to the research setting thanks to validation, regulatory approval, payer coverage, and cost issues, the researchers cautioned.
Going forward, artificial intelligence (AI) can be used to enable MRD assessment. The tool can analyze large amounts of data and identify patterns human reviewers may not detect. AI can also improve the speed and accuracy of MRD measurements, simplify the interpretation of molecular testing, and reduce errors related to operator dependency.
In addition, AI holds the potential to yield recommendations for a personalized precision medical decision-making algorithm.
However, “as MRD assessment via liquid biopsy in myeloid malignancies remains under investigation, very little research exists on the use of AI tools in the assessment of MRD via information collected by liquid biopsies,” the authors noted. “Current ongoing and future research will make this field more promising.”
Taken as a whole, the published literature includes several clinical studies that incorporate blood-based LBs using different techniques for MRD assessment, they concluded.
1. Allam S, Nasr K, Khalid F, et al. Liquid biopsies and minimal residual disease in myeloid malignancies. Front Oncol. Published online May 5, 2023. doi:10.3389/fonc.2023.1164017
2. Belkhair J, Raissi A, Elyahyaoui H, Ameur MA, Chakour M. Atypical chronic myeloid leukemia BCR-ABL 1 negative: a case report and literature review. Leuk Res Rep. Published online May 25, 2019. doi:10.1016/j.lrr.2019.100172