ProGraME : A novel flow cytometry algorithm for the diagnosis of low‐risk myelodysplastic syndromes in patients with cytopenia

• Published in: European journal of haematology Vol. 111; no. 6; pp. 851 - 862
• Main Authors: Therkelsen, Jesper, Træden, Dicte Wilhjelm, Schjødt, Ida, Andersen, Mette Klarskov, Sjö, Lene Dissing, Hansen, Jakob Werner, Grønbæk, Kirsten, Dimopoulos, Konstantinos
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc 01.12.2023
• Subjects: Algorithms; Bone dysplasia; Bone marrow; CD34 antigen; CD36 antigen; Cell lineage; Cytogenetics; Cytopenia; Diagnosis; Flow cytometry; Gating; Leukocytes (granulocytic); Lymphocytes; Monocytes; Myelodysplastic syndrome; Progenitor cells; Standardization
• ISSN: 0902-4441; 1600-0609
• DOI: 10.1111/ejh.14086

Abstract Objectives Flow cytometry (FC) is, together with morphology, genetics, and cytogenetics, used in the diagnostic assessment of cytopenia, as its value in evaluating bone marrow dysplasia been highlighted by several studies. However, despite the development of algorithms and guidelines, there is still a lack of standardization of the FC assessment of bone marrow dysplasia. Methods By combining FC, together with morphological analysis and cytogenetic/molecular assessment in a training cohort of 209 patients, we created a novel score, ProGraME, which includes four parameters, each from a different cell lineage (Progenitor cells, Granulocytes, Monocytes, Erythroid precursors), solely based on relevant population gating. Points for ProGraME were attained for: lymphoid precursors ≤5% of all CD34+ cells (1.5 point); a granulocyte‐to‐lymphocyte side‐scatter ratio ≤6 (1 point); a monocyte CD33‐CV% ≥ 63 (2 points), and an erythroid precursor CD36‐CV% ≥ 65 (2 points). Results Using a cutoff of ≥2 as suggestive of dysplasia, ProGraME showed a sensitivity of 91% and a specificity of 81% in the training cohort and 95% and 75%, respectively, in an independent validation cohort of 159 patients. In addition, ProGraME had a very high negative predictive value of 97.1% and 97.8% in the training and validation cohorts, respectively, offering a useful tool for excluding bone marrow dysplasia. Finally, among the 23 CCUS patients that scored positive for dysplasia with ProGraME in the training cohort, 16 of them (69%) carried high‐risk mutations, suggesting that FC might help identify early changes of dysplasia. Conclusions ProGraME can potentially optimize the FC diagnosis of low‐risk myelodysplasia without minimal requirements of flow analysis other than accurate population gating.