B-ALL MRD After Anti-CD19 Therapy: 27-Color Flow Cytometry and Stage-0 Hematogones

What COMIP-002 reports

Shean and colleagues describe a 27-colour spectral flow cytometry panel for bone marrow measurable residual disease (MRD) assessment in B-cell acute lymphoblastic leukaemia after treatment, including anti-CD19 and anti-CD22 therapies. Rather than relying on CD19 alone, their pan-B lineage strategy uses a Boolean combination of CD19, CD72, CD24 and CD22. A particularly important target is the normal CD19-negative stage-0 haematogone: CD34-positive, CD22-positive and CD19-negative, with bright CD81 and CD38 expression. Such precursors can appear in regenerating marrow and resemble residual leukaemic blasts. During development, 11 positive and ten negative samples compared against an existing 10-colour assay were reportedly concordant. The paper states that formal validation was not complete and that clinical sensitivity and the lower detection limit had not been established. These 21 comparisons cannot establish a generalisable limit of detection or substitute for local clinical validation. Primary source: Shean et al., Cytometry Part B (2026).

When treatment changes your familiar gate

Antigen-directed therapies can alter surface expression on residual malignant cells. A method that begins by excluding CD19-negative events may lose the very population that needs investigation. Broadening lineage search is not equivalent to relaxing one fluorescence threshold: every additional anchor captures normal cells that must be distinguished from disease. Predefine the rule for discordant lineage markers and preserve an audit trail for inclusion and exclusion decisions. The search should remain inclusive enough to detect possible escape phenotypes and specific enough to avoid calling normal precursors malignant.

Stage-0 haematogones: specificity requires developmental context

The normal CD34/CD22/CD19/CD38/CD81 relationship is part of the diagnostic context. Do not classify an isolated CD19-negative cluster as MRD solely because it lies outside a conventional CD19-positive B-cell gate. Assess normal B-cell maturation, CD9, CD24, CD44 and other differentiation-associated and aberrant antigens, comparing with appropriate post-therapy and regenerating marrow controls. The most informative display for one specimen may be misleading in another. A final dot plot is an illustration of a multiparameter judgment, not all of its evidence.

Denominator, counting and confirmation

Document viable-event selection, doublet removal, debris exclusion, time gating and denominator choice. The COMIP-002 workflow reports events over viable leukocytes; this is not interchangeable with reporting over viable mononuclear cells. Rare candidate populations should be assessed for event count, phenotypic coherence, acquisition artefacts and stability of gate placement. Retain the raw FCS and analysis versions for review. A very small population may be detected below a validated quantification boundary, which must be reported as such, rather than silently treated as either a robust positive or a negative.

Clinical validation remains a separate experiment

Before implementation, evaluate analytical background, detection and quantification limits, dilution recovery, reproducibility, operator differences and therapy-era case mix, including CD19-low and CD19-negative specimens. Plan comparisons with an established clinical MRD workflow and orthogonal resolution of discordant cases. A development-panel description and agreement in 21 selected cases do not establish sensitivity across laboratories, cytometers or treatment stages. The assay's intended specimen, interpretation and clinical decision boundaries should be written before declaring it ready.

Daily Learning Pill — one antigen is not a lineage

After CD19-directed treatment, absence of CD19 does not imply absence of malignant B-lineage cells. Search more broadly, then prove that the suspicious events differ from physiological regeneration. A negative result is meaningful only for the populations the assay could reliably detect.

Related CFCM learning and next step

See the CFCM library for flow cytometry panel design and validation resources. In your next B-ALL MRD review, document whether the initial candidate search would retain a CD19-negative disease population.

Evidence boundary

This Technical Review is an original CFCM educational interpretation of the paper's Purpose, Background, Table 2 and Gating Strategy. The supplements and patient-level data were not independently inspected. Reagents and examples in the paper do not imply that a new panel is clinically validated or approved. DOI 10.1002/cyto.b.70044.