Laboratory Aspects of Minimal / Measurable Residual Disease Testing in B-Lymphoblastic Leukemia

for B-lymphoblastic leukemia (B-ALL) in response to up front chemotherapy and success in bone marrow (BM) transplantation for relapsed and high-risk ALL. First shown in the pediatric population, similar usefulness has been shown for the adult population.

Over time and with the introduction of even more sensitive assays, the threshold for minimal is changing and MRD is being redefined as measurable residual disease.

MRD is often defined as disease below the level of reliable detection by morphology or 5% blasts of the total marrow cellularity.

Flow cytometry studies have shown that this 5% value is not very sensitive and values as low at 0.01% residual disease during treatment are predictive of relapse. In addition, high percentages (>5%) of normal precursor B cells or lymphoblasts can be seen in normal or regenerating marrow without residual leukemia,

or quantitative reverse transcriptase PCR for fusion transcripts in some leukemia with recurrent chromosomal rearrangement, ^, both with a sensitivity of 0.001%. More recent approaches have even higher level of sensitivity of 0.0001%. These approaches include next-generation sequencing (NGS) using linear PCR amplification of T-cell receptor or B-cell receptor rearrangements followed by sequencing and comparing the products to the signature leukemic sequences identified at diagnosis. This approach has a sensitivity of 1 cell in 1 million or 0.0001% and, in certain circumstances, residual disease at even less than 0.01% is predictive of outcome. ^, Digital droplet PCR, the latest version of quantitative PCR, is being evaluated in small studies with similar sensitivity to NGS. ^, Two studies have examined the clinical significance of MRD detected by flow cytometry compared with allelic-specific quantitative PCR or with NGS.

PCR, as expected with its better sensitivity, can detect residual disease in some of the flow MRD negative cases, but missed some detected by flow, possibly because of clonal evolution and sequence drift. NGS using 0.01% positivity cutoff detected all flow positive cases and identified additional cases that were negative by flow. Outcomes for these 2 studies were similar, relapse occurred most frequently when MRD is positive with both test modalities and least frequently when negative for both tests. In patients where the MRD was positive on only 1 testing platform (either flow- or nucleic acid-based MRD methodologies) an intermediate level of relapse (event-free survival) was observed compared with high- or standard-risk patients.

The remainder of this article focuses on the flow cytometry–based MRD detection for B-ALL that is used by many institutions. This modality is the most common method and has advantages of rapid turnaround time (MRD results reported usually within one working day compared with 7–10 days for the other modalities) and its availability in most clinical laboratories. There is no consensus on the timing, preparation, acquisition, and analysis of flow cytometry. Despite all the variability in flow-based assays, MRD findings in multiple published large studies are equivalent with similar levels of sensitivity and correlation to outcome.