A Simple Approach to Flow Cytometric Assessment of Myeloid Dysmaturation
Dragan Jevremovic, Michael T Timm, Curtis A Hanson, William G Morice, Phuong L Nguyen. Mayo Clinic, Rochester, MN
Background: Flow cytometry immunophenotyping (FCIP) has been suggested as an adjunctive technique in the diagnosis of myeloid malignancies, including myelodysplastic syndromes (MDS), myeloproliferative neoplasms (MPN) and myelodysplastic/myeloproliferative neoplasms (MDS/MPN). However, there are significant challenges with FCIP diagnostic approach, including lack of uniformity of testing between laboratories, poor reproducibility of results, and high complexity of analysis with 2-dimensional displays. In this study, we have attempted to distinguish myeloid neoplasms from reactive bone marrow conditions using a simplified FCIP analysis of bone marrow blasts.
Design: Bone marrow aspirates from 32 patients with myeloid neoplasms and 9 normal controls were examined by FCIP using a single 8-color tube with antibodies to CD13, CD15, CD16, CD33, CD34, CD45, CD117, and HLA-DR. 300,000 events were collected per case. The data was analyzed using BD FACSDiva software. Blasts were gated using CD45/side scatter and CD34/CD117 plot. CD34+CD117+ cells were displayed on CD13/HLA-DR plot.
Results: Out of 32 patients with myeloid neoplasms, there were 18 with MDS (6 of whom had <5% blasts), 5 with MPN, and 9 with MDS/MPN (including 5 patients with chronic myelomonocytic leukemia). 9 normal controls included 4 patients with left-shifted granulocytic hyperplasia due to growth factor therapy.
All samples from normal controls, including those with left shifted granulopoiesis, showed scattered blasts on CD13/HLA-DR plot, with 3 readily identifiable subpopulations: CD13brightHLA-DRbright, CD13moderateHLA-DRdim, and CD13dimHLA-DRbright. In contrast, the majority of patients with MDS (16 of 18), MPN (3 of 5), and MDS/MPN (7 of 9) had an abnormal CD13/HLA-DR expression on the CD34+CD117+ blasts. The most common abnormality was concentration of blasts in a single cluster, with the concomitant loss of normal scatter and subpopulations.
Conclusions: While there are many multicolor FCIP approaches for detection of myeloid abnormalities in the bone marrow, they all rely on complex interpretation algorithms, with or without help of a specifically designed software. Our findings provide a possible alternative, using a relatively simple gating strategy and a limited number of antibodies for a clear 2-dimensional visual display. Although some of the sensitivity is lost by omitting large number of antibodies for analysis, this method is potentially clinically useful, as it showed high specificity in distinguishing patients with myeloid neoplasms with <5% blasts from patients with left-shifted granulopoiesis due to growth factor exposure.
Tuesday, March 20, 2012 9:30 AM
Poster Session III # 227, Tuesday Morning