CD38 Expression On CD34+ Cells as a Single Flow Cytometric Parameter to Distinguish Myelodysplastic Syndromes (MDS) From Non-Clonal Cytopenias.

Abstract 2783

Poster Board II-759

Myelodysplastic syndromes are a heterogeneous group of bone marrow (BM) disorders characterised by clonal proliferation. The diagnosis of MDS can be challenging, especially in cases with minimal dysplastic changes by morphology and without characteristic karyotypic abnormalities. Recent investigations show that multi parameter analysis of BM samples of MDS patients by flow cytometry (FCM) might be of value in the diagnosis of MDS and for discriminating MDS from other disease entities. Detection of immunophenotypic abnormalities in the immature and the maturing myelomonocytic compartment is even included as a co-criterion in the minimal diagnostic criteria for MDS (Valent et al., Leuk Res 2009,31:727–36). Integration of FCM as a diagnostic tool in MDS requires a simple method with high sensitivity and specificity with low inter laboratory variation. Recently, reduced CD38 expression was demonstrated to discriminate between MDS and other disease entities (Goardon et al., Haematologica 2009, 94:1160–63).To validate the value of CD38 in the diagnosis of MDS, we analysed BM samples of 27 MDS patients (refractory anemia n=4; refractory anemia with multi lineage dysplasia n=15; refractory anemia with excess blasts [RAEB]-1 n=4 and RAEB-2 n=4), age-matched healthy volunteers (n=16) and pathologic controls (n=39). The pathologic control group consisted of aplastic anemia (n=3), iron deficiency (n=6), Multiple Myeloma (n=8), B-cell non-Hodgkin lymphoma (n=6) and autoimmune thrombocytopenia (n=3) patients. We found that CD38 RMFI on CD34+ cells was significantly decreased in MDS with less than 5% blasts (mean±SD: 216.5±95.4) as compared to pathologic controls (mean±SD: 377.2±246.1, P=0.01) and healthy volunteers (mean±SD: 322.8±138.1, P=0.02). In RAEB-1/-2, CD38 RMFI (mean±SD: 111.3±74.9, P=0.001) was even more decreased in comparison with both pathologic controls and healthy volunteers. Expression of CD38 on CD34+ cells is heterogeneous and the most immature cells or hematopoietic stem cells are considered to be negative for CD38 expression. By using the median fluorescence intensity (RMedFI), the skewed distribution of CD38 expression on CD34+ cells is taken into account. In our cohort, RMedFI is significantly decreased in MDS with <5% blasts (mean±SD: 368.1±210.3, P=0.03) and MDS RAEB -1/-2 (mean±SD: 139.7±108.5, P<0.001) as compared to pathologic controls (mean±SD: 521.6±314.2). However, no significant differences were observed between RMedFI of CD38 of MDS with less than 5% blasts and healthy volunteers. In conclusion, we show that CD38 expression is reduced in CD34+ cells of MDS patients and might be of value in the diagnosis of MDS. However, there is considerable overlap in CD38 expression between the MDS with less than 5% blasts group and pathologic controls group and healthy volunteers group. Differentiation of MDS from other disease entities by FCM is based on an accumulation of aberrancies rather than a single aberrancy which might reflect the degree of deviation from normal. The number of aberrancies in the myelomonocytic and blast compartment can be translated by a flow cytometric scoring system into a flow score (Wells et al., Blood, 2003,102:394–403, van de Loosdrecht et al., Blood, 2008,111:1067–77). A flow score above 3 is suggestive for MDS. The clinical impact of CD38 expression as a single flow cytometric parameter or as part of a multi parameter analysis requires validation with emphasis on inconclusive cases such as idiopathic cytopenia of unknown significance and idiopathic dysplasia of unknown significance. Flow cytometry in MDS is highly promising and prospective studies are warranted.