Dim CD33 Expression in Myelomonocytic Marrow Elements: A Novel Description of an Apparently Normal Variant
Shannon K Rathke, Alexandra M Harrington, Horatiu Olteanu, Steven H Kroft. Medical College of Wisconsin, Milwaukee, WI
Background: CD33 expression is evaluated frequently by flow cytometry (FC) in the work-up of myeloid disorders. We have anecdotally observed patients with dim CD33 expression in all myelomonocytic cells in repeated FC studies. To our knowledge, this phenomenon has not been previously reported. We sought to specifically evaluate this phenomenon and determine its prevalence in our patient population.
Design: All patients with more than one serial FC evaluation of marrow using panels containing CD33 from 2005-2012 were initially included in the study. Patients in whom all examined marrows were positive for a myeloid neoplasm were excluded. CD33 expression was semi-quantitatively classified. Myelomonocytic elements were considered dim for CD33 if the maximal CD33 expression on monocytes was one log below cases with normal expression.
Results: Of 422 patients evaluated, 25 (5.9%) showed consistently dim CD33 expression in multiple FC studies (2-12 studies, median 4). In all cases, there was a proportionate decrease in CD33 expression on blasts, granulocytes, and monocytes, such that the relative relationships between levels of CD33 expression in these populations was maintained. In the CD33(dim) group, there were 12 males and 13 females, with ages ranging from 19-71 (median 50). Primary diagnoses in these 25 patients included AML (17), biphenotypic acute leukemia (2), B-ALL (2), hemolytic anemia (1), cartilage hair-hypoplasia syndrome (1), CLL (1), and AML arising from CML (1). Nine of 25 patients underwent allogeneic stem cell transplant (ASCT), after which 6 showed normal CD33 expression, while 3 remained dim. Each of the 3 ASCT patients who remained CD33(dim) received transplants from related donors.
Conclusions: Dim CD33 expression on myelomonocytic cells appears to be a normal variant, seen in 5.9% of our patient population undergoing more than one FC study including CD33. This finding does not appear to be due to technical artifact, as the finding is reproducible over time in multiple FC studies of multiple patients. Furthermore, it can be corrected in association with ASCT. The finding is consistent across all stages of granulocyte and monocyte differentiation, possibly suggesting a decreased number of CD33 antigens on the cells or decreased affinity of the anti-CD33 for an altered antigen. It is important to be aware of this phenomenon so as to not mistake this apparently constitutional diminished CD33 expression for a marker of myeloid neoplasia.
Tuesday, March 5, 2013 9:30 AM
Poster Session III # 175, Tuesday Morning