Flow Cytometric Characteristics of CD34+ Cells in Refractory Anemia: Their Diagnostic Value.

Abstract

(INTRODUCTION) The diagnosis of refractory anemia (RA), a subtype of myelodysplastic syndromes (MDS), is often problematic because RA often does not show cytogenetic abnormalities and dysplastic hematopoietic cells are not specific to MDS. Recent studies have suggested that the abnormalities of hematopoietic cells detected by flow cytometry (FCM) might be useful for MDS diagnosis. However, this concept has not yet been widely accepted. We reported that MDS blasts often show aberrant antigen expression. In this study, we examined the diagnostic utility of the flow cytometric characteristics of CD34+ cells in RA. (METHODS) Whole bone marrow cells obtained from RA patients and controls (e.g., cytopenic individuals other than RA and other malignancies) were analyzed using four-color FCM. The cells were stained with CD45 and CD34, for gating CD34+ B-cell precursors (hematogones) and CD34+ myeloblasts, in combination with various other antibodies. We determined percentages of hematogones and CD34+ myeloblasts and relative mean florescence intensity (RMFI: MFI of each antigen staining ÷ MFI of isotype-matched negative control staining) of various antigens on CD34+ myeloblasts. (RESULTS) From the data on 50 control subjects in the first subject cohort, we determined the reference range (RR) for each flow cytometric parameter. We scored data on each subject by giving 1 point to each of 10 parameters if the data of such parameters were outside RRs. These 10 parameters, which were adopted after comparing data from control and RA subjects in the first cohort, were percentages of hematogones and CD34+ myeloblasts and RMFI of CD4, CD56, CD11b, CD13, CD15, CD33, CD117 and CD133. In the first cohort, scores of all control subjects were less than 3 (40 subjects had a score of 0) and scores of 7/12 RA patients were 3 or higher. Using the same parameters and same RRs, we then prospectively analyzed the second subject cohort. Again, this approach identified 8/13 RA patients (score 3 or higher) with 100% specificity. The degree of dysplasia and presence or absence of cytogenetic abberations were not related to RA identification. (CONCLUSION) Our method is reproducible and helpful in RA diagnosis.