Multiparameter Flow Cytometry as a Powerful Tool

Minimal residual disease (MRD) detection following therapy is rapidly becoming the standard of care for patients with acute leukemia. Flow cytometry is one of the principal technologies employed for MRD monitoring due its rapid turnaround time, similarity to diagnostic immunophenotyping, and widespread availability of instrumentation in clinical laboratories. The immunophenotypic identification of neoplastic cells relies on the altered expression of proteins or antigens by neoplastic cells in comparison to their normal counterparts of similar lineage and maturational stage, and represents an integration of alterations in the underlying genetic program that are responsible for neoplasia. However, important technical and interpretative differences exist between diagnostic immunophenotyping and MRD monitoring that require special attention to ensure consistent quality, not the least of which is appropriate assay validation. In addition, there is also significant variability in the methods of sample preparation, reagent panels used, and data analysis strategies employed in clinical flow cytometry laboratories and this variability extends to MRD testing. The consequence is variability in quality between laboratories for MRD testing, even laboratories considered knowledgeable or experienced. This lack of standardization represents a major challenge to widespread implementation of MRD monitoring by flow cytometry. Consequently, on-going efforts by the Childrens Oncology Group and Foundation of the National Institute of Health are focused on improving standardization of B lymphoblastic leukemia MRD testing for clinical trials and ultimately all patients. Nevertheless, in experienced laboratories using a standardized protocol, MRD monitoring by flow cytometry in B lymphoblastic leukemia has been shown to be capable of producing reproducible results at both the technical level and relative to patient outcomes on clinical trials, emphasizing that the technology itself allows for reproducible testing and that the current variability is largely due to implementation details and interpretive skills. Whether this degree of comparability can be achieved for T lymphoblastic leukemia or acute myeloid leukemia remains to be demonstrated. A related issue is the level of sensitivity that is both achievable and desirable for MRD monitoring at particular time points after therapy. Current flow cytometric assays have a routine limit of detection at early time points after therapy of roughly 0.01% of nucleated cells for acute lymphoblastic leukemia and 0.1% for acute myeloid leukemia with a higher sensitivity possible for subsets of patients having either more frankly aberrant immunophenotypes or reduced background populations of similar immunophenotype. Interestingly, our recent comparison of flow cytometry with high throughput sequencing in B lymphoblastic leukemia suggests that increased assay sensitivity does not necessarily improve risk stratification at early time points and that the moderate sensitivity of current flow cytometric assays is adequate for this purpose. However, higher sensitivity assays will likely be more important at later time points further from therapy. Multiple studies have now demonstrated the prognostic significance of MRD detection in acute leukemia after therapy or prior to bone marrow transplantation using flow cytometry. This raises the possibility as to whether the assessment of MRD after therapy can be used not only to assess response to therapy, but as a surrogate biomarker for outcome to expedite the new drug approval process.