Multiparameter flow cytometry reveals myelodysplasia-related aberrant antigen expression in myelodysplastic/myeloproliferative neoplasms


  • How to cite this article: Kern W, Bacher U, Schnittger S, Alpermann T, Haferlach C, Haferlach T. Multiparameter flow cytometry reveals myelodysplasia-related aberrant antigen expression in myelodysplastic/myeloproliferative neoplasms. Cytometry Part B 2012; 84B: 194–197.



Within the myelodysplastic/myeloproliferative neoplasm (MDS/MPN) category of the WHO (2008), only chronic myelomonocytic leukemia was so far evaluated by multiparameter flow cytometry (MFC).


To investigate the potential of MFC for MDS/MPNs, unclassifiable (MDS/MPNu), and refractory anemia associated with ring sideroblasts and marked thrombocytosis (RARS-T), we studied 91 patients with these entities (60 males/31 females; 35.3–87.4 years) for MDS-related aberrant immunophenotypes (≥2 different cell lineages with ≥3 aberrantly expressed antigens). Data were correlated with cytomorphology and cytogenetics.


MFC identified MDS-related immunophenotypes in 54/91 (59.3%) of patients. Patients with or without MDS-related immunophenotype did not differ significantly by demographic characteristics, blood values, or median overall survival. MDS-related immunophenotype cases showed a higher number of aberrantly expressed antigens (mean ± SD, 4.9 ± 2.4 vs. 2.0 ± 1.4; P < 0.001). Aberrant karyotypes showed a similar frequency in patients with and without MDS-related immunophenotype (11/54; 20.4% vs. 7/37; 18.9%; P = n.s.).


MDS-related immunophenotype are present in more than half of patients with MDS/MPNu and RARS-T. MFC therefore may be helpful to separate cases into more “MDS-like” or “MPN-like” subgroups. © 2012 International Clinical Cytometry Society

Immunophenotyping by multiparameter flow cytometry (MFC) is increasingly used in the diagnostic work-up of patients with cytopenias and suspected myelodysplastic syndromes (MDS). Myelodysplastic/myeloproliferative neoplasms (MDS/MPN) comprise a group of disorders in which the clinical, morphologic, and laboratory findings vary along a continuum from MDS to MPN (1). They define a separate category in the current WHO classification (2). The immunophenotypes of MDS (3) and chronic myelomonocytic leukemia (CMML) (4–6) have been described in detail: For example, MDS cases frequently show a reduced sideward scatter (SSC) signal or aberrations of the CD11b/CD16 and CD13/CD16 patterns of the granulopoietic cells (7), or abnormal CD71 expression of erythropoietic cells, and CMML features aberrant CD2 and CD56 expression on the monocytic cells. Flow cytometric scoring systems were even shown to be prognostically relevant in patients with MDS (8–10). In contrast, data are limited on the use of MFC in MDS/MPNs, unclassifiable (MDS/MPNu) as well as on refractory anemia associated with ring sideroblasts and marked thrombocytosis (RARS-T). MDS/MPNu includes patients that show at the time of initial presentation clinical, laboratory, and morphological features that overlap both MDS and MPN but do not meet the criteria for a distinct entity such as CMML. RARS-T is a provisional entity in the 2008 WHO classification which is defined by the presence of dysplastic, ineffective erythroid proliferation, ring sideroblasts ≥15% of erythroid precursors in the bone marrow, and thrombocytosis ≥450 × 109/l (2).

Aiming to increase insights in the potential of immunophenotyping for these entities, we investigated 91 patients with MDS/MPNu and RARS-T for MDS-related aberrant immunophenotypes and performed correlation studies with other diagnostic techniques as patients were investigated by a comprehensive diagnostic work-up including cytomorphology and cytogenetics. Eighty-five patients had MDS/MPNu and six patients had RARS-T according to WHO definition (2). There were 60 males and 31 females at a median age of 75.1 years ranging from 35.3 to 87.4 years. Bone marrow samples were sent for diagnosis from different hematological centers to the MLL Munich Leukemia Laboratory between August 2005 and June 2011. All patients gave their written informed consent to genetic analyses and the performance of research studies. The study was approved by the Internal Review Board of the Munich Leukemia Laboratory and followed the Declaration of Helsinki. All 91 cases were analyzed by bone marrow cytomorphology combined with iron staining (11), chromosome banding analysis using standard methods combined with fluorescence in situ hybridization (FISH) if needed (12), and immunophenotyping by five-color staining (10).

For MFC, samples underwent ammonium chloride-based erythrocyte lysis for reduction of the number of erythrocytes. Subsequently, 5-color staining was performed using antibody panels for detection of MDS described before (10). Immunophenotyping was performed on FC500 and Navios flow cytometers (Beckman Coulter, Miami, FL), and data analysis was done by CXP and Kaluza software packages (Beckman Coulter). About 50,000 events were evaluated in each bone marrow sample. Briefly, myeloid blasts were evaluated for coexpression of CD11b, CD5, CD7, CD56, CD15, and CD64, and HLA-DR negativity. Granulocytes were evaluated for reduced SSC signal, abnormal CD11b/CD16 and CD13/CD16 expression patterns, CD56 coexpression, and CD33 and CD64 negativity. Monocytes were evaluated for reduced expression of CD11b, HLA-DR, CD13, coexpression of CD16, CD56 overexpression, and aberrant CD2 coexpression. Erythroid cells were evaluated for CD71 expression. The immunophenotype was rated MDS-related if in ≥2 different cell lineages a total of ≥3 aberrantly expressed antigens were found (10). Dichotomous variables were compared using chi-square test, continuous variables by Student's T-test (differences in variances were analyzed by Levene-test). Overall survival (OS) was calculated by Kaplan–Meier method and compared by log-rank test. SPSS software (version 19.0, IBM Corporation, Armonk, NY) was used.

MFC identified an MDS-related immunophenotype in 54/91 (59.3%) patients, in detail in 50/85 (58.8%) cases with MDS/MPNu and in 4/6 (66.7%) cases with RARS-T. Patients with or without MDS-related immunophenotype did not differ significantly by mean age, sex distribution, or mean peripheral blood values (Table 1). Cases with MDS-related immunophenotypes displayed aberrant antigen expression significantly more frequently than those without as follows: myeloid progenitor cells (mean number of aberrantly expressed antigens ± SD), 0.5 ± 0.6 vs. 0.2 ± 0.4; P = 0.002; granulocytes, 2.7 ± 1.3 vs. 1.2 ± 1.1; P < 0.001; monocytes, 1.7 ± 1.2 vs. 0.5 ± 0.7; P < 0.001; erythroid cells, 0.3 ± 0.4 vs. 0.1 ± 0.3; P = n.s.). Therefore, the MDS-related immunophenotype cases expressed a significantly higher mean total number of aberrantly expressed antigens (4.9 ± 2.4 vs. 2.0 ± 1.4; P < 0.001) (Table 1). Table 1 provides the respective details on all antigens for which significant differences in aberrant expression were found between cases with and without an MDS-related immunophenotype. Prognosis did not differ significantly between both subgroups as patients with an MDS-related immunophenotype had a median OS of 42.3 months, whereas in those without an MDS-related immunophenotype the median OS was not reached (P = n.s.; Table 1).

Table 1. Comparison of Demographic Data, Peripheral Blood Values, the Number of Aberrantly Expressed Antigens by Multiparameter Flow Cytometry, and of Median Overall Survival (OS) of Patients With and Without an MDS-Related Immunophenotype (10)
ParameterTotal cohort n = 91 (100.0%)MDS-related immunophenotype n = 54 (59.3%)No MDS-related immunophenotype n = 37 (40.7%)P-value
  1. WBC counts: white blood cells count; SD: standard deviation; n.s.: non-significant; OS: overall survival.

Demographic data    
 Males/females (sex ratio)60/31 (1.9)37/17 (2.2)23/14 (1.6)n.s.
 Mean age, years (±SD)72.2 (±10.9)72.8 (±10.1)71.2 (±12.0)n.s.
Peripheral blood values    
 Mean WBC count, ×109/l (±SD)27.5 (±27.4)30.9 (±30.2)22.2 (±21.7)n.s.
 Mean Hb, g/dl (±SD)10.2 (±2.6)10.3 (±2.6)10.1 (±2.7)n.s.
 Mean platelets, ×109/l (±SD)288 (±275)264 (±264)322 (±292)n.s.
Number of aberrantly expressed antigens (mean ± SD)    
 Myeloid progenitors0.4 (±0.5)0.5 (±0.6)0.2 (±0.4)0.002
 Granulocytes2.1 (±1.4)2.7 (±1.3)1.2 (±1.1)<0.001
 Monocytes1.2 (±1.2)1.7 (±1.2)0.5 (±0.7)<0.001
 Erythroid cells0.2 (±0.4)0.3 (±0.4)0.1 (±0.3)n.s.
 Total (all cell compartments added)4.0 (±2.5)4.9 (±2.4)2.0 (±1.4)<0.001
Frequencies of aberrancies in granulocytes    
 Reduced side scatter signal54 (59.3%)40 (74.1%)14 (37.8%)0.001
 Aberrant CD13/CD16 pattern48 (52.7%)39 (72.2%)9 (24.3%)<0.001
 Aberrant CD11b/CD16 pattern33 (36.3%)30 (55.6%)3 (8.1%)<0.001
Frequencies of aberrancies in monocytes    
 Aberrant CD56 expression45 (49.5%)34 (63.0%)11 (29.7%)0.003
 Reduced CD4 expression8 (8.8%)8 (14.8%)-0.019
 Reduced CD13 expression17 (18.7%)15 (27.8%)2 (5.4%)0.007
 Reduced HLA-DR expression20 (22.0%)18 (33.3%)2 (5.4%)0.002
Outcome analysis    
 Median OS (months)42.342.3not reachedn.s.

The majority of patients had normal karyotypes (73/91; 80.2%), and only 18 (19.8%) had aberrant karyotypes. Aberrant karyotypes showed a similar frequency in patients with and without MDS-related immunophenotype (11/54; 20.4% vs. 7/37; 18.9%; P = n.s.; Table 2).

Table 2. Frequency of Cytomorphologic Subtypes and of Aberrant Karyotypes in Patients With and Without an MDS-Related Immunophenotype (10)
ParameterTotal cohort n = 91 (100.0%)MDS-immunophenotypeNumber of aberrantly expressed antigens (for molecular mutations: with vs. without mutation)
Yes, n = 54 (59.3%)No, n = 37 (40.7%)P-value P-value
  1. MDS/MPNu: myelodysplastic/myeloproliferative neoplasm, unclassifiable; RARS-T: refractory anemia associated with marked thrombocytosis; SD: standard deviation; n.s.: non-significant; KT: karyotype.

Cytomorphologic subtype
 MDS/MPNu85 (100.0%)50/85 (58.8%)35/85 (41.2%)n.s.  
 RARS-T6 (100.0%)4/6 (66.7%)2/6 (33.3%)   
Chromosome banding analysis
 Aberrant karyotype18/91 (19.8%)11/54 (20.4%)7/37 (18.9%)n.s.4.8 ± 3.0n.s.
 Normal karyotype73/91 (80.2%)43/54 (79.6%)30/37 (81.1%) 3.7 ± 2.4 

These results demonstrate that the presence of dysplastic features in the MDS/MPN category by morphology finds its reflection on the immunophenotypic level as the majority of patients with MDS/MPNu and RARS-T show MDS-related aberrant antigen expression. If confirmed in larger series, this data may form a basis to subdivide MDS/MPNu based on molecular genetics and on the immunophenotype into cases with MDS-related features and those without. Further analyses are particularly needed to validate these findings and their potential significance in RARS-T since only six such cases were included in the present series. By SNP array karyotyping, Gondek et al. described that uniparental disomy and cryptic chromosomal abnormalities in MDS/MPNu were more frequent in JAK2V617 wild-type cases as compared to JAK2V617F positive cases (13). This also emphasizes the existence of different molecular genetic subgroups in the MDS/MPNu category. Finally, this report emphasizes to further investigate the potential of MFC for diagnosis of patients with MDS/MPNs. MFC may help to separate the MDS/MPNu category into more “MDS-like” or more “MPN-like” cases and therewith help to guide molecular analyses into the direction of MDS or MPN related mutations and eventually in the future toward a more specific treatment.

Authorship and Disclosures

WK and TH performed study design. WK was responsible for immunophenotyping, TH for cytomorphology, SS for molecular mutation analysis, and CH for cytogenetics. UB and WK performed data analysis and wrote manuscript. TA contributed to data analysis. All authors contributed to write the manuscript, reviewed the manuscript, and gave their consent to the final version. WK, SS, CH, and TH declare part ownership of the MLL Munich Leukemia Laboratory GmbH. UB and TA are employed by the MLL Munich Leukemia Laboratory GmbH.