Intelligence

Analysis

AML MRD: what to count inside the diagnostic gate

An AML follow-up gate can contain cells that lack the aberrancy that made it useful at diagnosis. Barone and colleagues compared counting the entire gate with applying an additional phenotypic check.

CFCM Technical Review of an analytical comparison in NPM1-mutated AML. DOI 10.3390/ijms26104506.

Standfirst: An AML follow-up gate can contain cells that lack the aberrancy that made it useful at diagnosis. Barone and colleagues compared counting the entire gate with applying an additional phenotypic check.

Two ways to read the diagnostic template

The authors compared two multiparameter flow methods with RT-qPCR for NPM1-mutated transcripts in 125 bone marrow samples from 25 people with NPM1-mutated AML. The first flow method counted the cells in a patient-specific template built at diagnosis. The LAIP-based different-from-normal method added a check for the aberrant markers of interest within that template. This distinction matters when the diagnostic phenotype is present in only part of the clone.

AML MRD: what to count inside the diagnostic gate (panel 1)
Original CFCM schematic of the two flow readouts compared with NPM1 RT-qPCR by Barone et al., International Journal of Molecular Sciences (2025), DOI 10.3390/ijms26104506.

Treatment groups gave different cutoffs

In the LAIP-based different-from-normal analysis, the reported ROC cutoffs were 0.034% after intensive chemotherapy and 0.095% in samples associated with venetoclax plus a hypomethylating agent. The comparator was molecular MRD. These are empirical classifications in the study population. They do not establish an analytical limit of detection or a universal threshold for treatment decisions.

Check what the comparator can answer

The study compared a phenotype-based count with a molecular readout in a small, mutation-defined cohort with repeated specimens. Before transferring the approach, verify the fraction of baseline cells bearing the aberrancy, how regenerating cells enter the gate, and how discordant molecular and flow results will be reviewed. Do not extrapolate the numerical cutoffs to AML without a trackable NPM1 mutation.

Daily Learning Pill

Record which diagnostic aberrancy makes the follow-up gate informative and how much of the baseline clone carries it.

Read the original article.