Blast counts are lower in the aspirate as compared to trephine biopsy in acute myeloid leukemia and myelodysplastic syndrome expressing CD56.

INTRODUCTION

CD56 is aberrantly expressed in myeloid neoplasms including myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Considering the adhesion effects of CD56, blast quantification in bone marrow might depend on the technique used to obtain respective diagnostic specimens. Therefore, the objective of our study was to investigate the impact of CD56-expression on blast counts in myeloid neoplasms comparing bone marrow aspirates to biopsies.

METHODS

We retrospectively analyzed 75 patients diagnosed with MDS and AML. We compared patients with (n = 36) and without (n = 39) CD56-expression by flow cytometry with respect to their blast quantities assessed on bone marrow aspirates versus biopsies.

RESULTS

The frequency of CD56-expression on blasts correlated with higher blast counts on biopsies vs. aspirate smears (r  = 0.52; P = .001). This difference in blast counts was only significant in the CD56 high expressing subgroup (median 68%, 5.5%-95% in biopsy compared to median 32.5%, 1.5%-90% in aspirate; P < .01). The percentage of CD56-positive blasts among the total blast population was lower in the peripheral blood compared to bone marrow (median 31%, 6%-88% vs. 55%, 14%-98%; P = .016). The discrepancy in the blast count between the aspirate and trephine biopsy would have led to misclassification of four cases as MDS instead of AML, if diagnosis had based on the bone marrow aspirate blast count alone.

CONCLUSION

Counting blasts in bone marrow aspirates of CD56-positive AML and MDS may be linked to underestimation, potentially leading to misclassification of these myeloid neoplasms, and should therefore be adjusted considering the results obtained on trephine biopsies for reliable diagnosis.