Abstract 2554


The Acute Myeloid Leukemias of Normal Karyotype (AML-NK) are a subgroup of hematologic malignancies with high heterogeneity, having some of them genetic markers with prognostic value. There is little information about the potential influence of different genetic markers in AML-NK on the immunophenotype of blast cells, particularly of immature populations.


The aim of the present study is to analyze the potential influence of the presence or absence of different genetic markers (FLT3-ITD, mutNPM1) and gene expression levels of BAALC, ERG, EVI1, MN1, PRAME, FOXO3A and WT1) on the immunophenotype of AML-NK, focusing on the phenotypic characteristics of immature blast cells (CD34+ cell compartment).

Material and methods:

We included in the study 57 cases of de novo AML-NK confirmed by conventional cytogenetics and absence of known molecular alterations by FISH (chromosomes 5,7,8, and 11q23 probes), in which extensive molecular studies as well as a large panel of four colour multiparametric flow cytometry immunophenotype (CD34, CD45, CD19, cyCD79a, nTdT, cyMPO, cyCD3, CD7, CD2, CD56, CD65, 7.1, CD11b, CD13, HLADR, CD117, CD61, CD33, CD123, CD36, CD64, CD14, CD71, Glycophorin A, CD15, CD16) were performed at diagnosis. The CD34 and CD45 antigens were included in all four colour combinations.

In 15/57 patients the FLT3 gene and 28/57 cases the NPM1 were mutated. The fold change in expression of the target genes relative to an internal control gene (ABL) was expressed as: Fold Change (RQ): = 2(–ΔCt sample) / 2(–ΔCt calibrator), where the DCt (Ct target gene – Ct ABL) of each patient sample is compared with a calibrator sample (mean expression of five bone marrow from healthy donors), using the DDCt method.

The median gene expression was 1.228, 0.392, 3.253 and 0.758 for BAALC, FOXO3A, MN1 and ERG and the 75th percentile was 0.361, 33.87 and 259.77 for EVI1, PRAME and WT1 respectively.

The threshold to define positivity for an antigen was set at 20% of blasts population analyzed. In addition, the complete immunophenotype was analyzed on the CD34+ blast cells when we identified a CD34+ blast cell subpopulation greater than 3% of the total blast cells, in order to include the minor subpopulations. The software SPSS v19 was used for the statistical analysis.


Upon comparing FLT3-ITD+ vs FLT3-ITD- patients, no differences were observed in the overall incidence of CD34+ and/or CD117+ cases. Nevertheless, if we considered cases with small CD34+ subsets (3%–20% of CD34+ blast cells) the frequency was higher among FLT3-ITD+ patients (40% vs 4%). Upon analyzing the phenotypic profile of the CD34+ cells according to FLT3-ITD status, in patients positive for this marker the CD34+ cells more frequently expressed CD7 (p=0,02) and CD36 (p=0,04).

The same comparison was done according to NPM1 status (mutated vs non mutated). The incidence of CD34+ cases was significantly lower while the number of CD117+/CD34- blasts was significantly higher in the mutated subgroup. No differences were observed in the phenotypic profile of the CD34+ cells.

For BAALC and MN1 expression, patients were divided according to the median value into two categories: high and low expression. In both markers, the high expression was associated with increased frequency of CD34+ cases (75 vs 17%, p=0,00) and (71 vs 21%, p= 0,01), respectively. The phenotypic profile of CD34+ cells did not differed between high and low groups.

Patients with high ERG expression showed a tendency to be associated with increased frequency of CD34+ cells, but differences did not reached statistical significance (p=0,08). Of note, CD34+ cells expressed cMPO less frequently among the ERG high patients.

We have no detected differences in the expression of blasts according to EVI1 or FOXO3A expression. Interestingly, in cases with low level of expression of these two markers, the CD34+ blast cell compartment more frequently expressed CD7 (p= 0,01 and p= 0,02, respectively).

Finally, PRAME and WT1 expression were not related with the CD34 expression neither with the phenotypic profile of the CD34+ blasts. However, high WT1 was associated with a significantly higher number of CD117+/CD34- cells (p<0.01).


The presence of FLT3-ITD and NPM1 mutations as well as the level of expression of relevant genes in AML-NK patients correlates with the CD34 expression on blast cells as well as the phenotypic profile of this immature cell subset.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.