Only recently, a class of gene regulatory molecules (miRNA’s) has been described that have the potential for posttranscriptional and translational gene regulation. MiRNA’s can influence critical cellular properties like proliferation and differentiation. Since these properties vary substantially in genetically defined AML subsets we have chosen to analyze AML patients with inv(16) (n=12); standard risk cytogenetic classification (mainly comprising the normal karyotype) (SR) (n=21) and with monosomy 7 or deletion7q (n=17) using DNA oligonucleotide arrays which are employing the mirVANA miRNA probe set. Principal Component Analysis (PCA) of AML samples from patients with inv(16) showed only minor differences to normal bone marrow (BM). In contrast AML patients with SR could be clearly divided in two subgroups - one subgroup that showed miRNA expression pattern close to BM and a distinctly different subgroup. Comparison of SR patient samples with BM using ANOVA tests revealed 9 differentially expressed miRNA’s (mir-223; mir-15a; mir-16; mir-203; mir-103; mir-23b; mir-107, mir-17–5p and mir23a). AML patients with aberrations of chromosome 7 showed highly distinct PCA pattern as compared to BM samples. 64 miRNA’s were found to be differentially expressed in those patients as compared to BM. All miRNA’s located on chromosome7 that were included in the array were found downregulated in AML patients with aberrant chromosome 7 (e.g.mir-335).

To further substantiate the microarray data we analyzed a total of 108 AML patients with inv(16), SR cytogenetics and aberrations of chromosome 7 for mir-23a, mir-223, mir-16 and mir-335 expression by qRT-PCR. Supporting the microarray data it could be shown that the median expression levels of mir-23a, mir-223 and mir-16 were lowest in patients with SR cytogenetics, whereas AML patients with chromosome 7 aberrations showed lowest median expression for mir-335 transcripts. This was in clear contrast to the expression of mir-150, a miRNA which could be previously shown to associate with immature T-cells and which had the highest expression in AML samples with chromosome 7 aberrations. Expression of mir-150 was correlated with a high CD34 percentage in the investigated AML samples (p<0.001) whereas mir-223 did not correlate with CD34 but was strongly associated with the surface expression of CD14 (p<0.001). In an in-vitro model of G-CSF induced CD34+ stem cell differentiation mir-223 and mir-16 were slightly upregulated whereas mir-150 disappeared during the process of differentiation. Finally, the investigated miRNA were entered in a model for outcome prediction in AML samples with SR cytogenetics including known risk factors. Within this model the ratio of mutant vs. wt FLT-3 was the strongest predictor of overall survival (p<0.01). Additonally, mir-23a expression proved to be an independent negative prognostic factor for AML patients with SR cytogenetics (p<0.05).

In conclusion, the presented data demonstrate specific expression profiles of miRNA’s in AML patients with different cytogenetic risk profiles. As it could be shown for patients with SR cytogenetics the observed expression profiles are likely to contribute to a deregulated differentiation and may influence therapeutic outcome.

Disclosures: D. Ovcharenko and D. Brown are employees of Asuragen Inc.; Research support for microarray analysis by Ambion/Asuragen Inc.

Author notes


Corresponding author