Overexpression of BRE in Pediatric MLL-Rearranged Acute Myeloid Leukemia Associated with t(9;11)(p22;q23).

Abstract 3471

Poster Board III-359

One important cytogenetic subgroup of pediatric acute myeloid leukemia (AML) is characterized by translocations of chromosome 11q23, which accounts for 15 to 20% of all cases with an evaluable chromosome analysis. In most of these cases, the mixed lineage leukemia (MLL) gene is involved. More than 50 fusion translocation partners of the MLL gene have been identified and outcome differs by translocation partner, suggesting differences in the biological background. So far these biological differences have not been unravelled. Therefore, we investigated the gene expression profiles of MLL-rearranged subgroups in pediatric AML in order to discover and identify the role of differentially expressed genes.

Affymetrix Human Genome U133 plus 2.0 microarrays were used to generate gene expression profiles of 257 pediatric AML cases, which included 21 pediatric AML cases with t(9;11)(p22;q23) and 33 with other MLL-rearrangements. With these profiles, we were able to identify a specific gene expression signature for t(9;11)(p22;q23) using an empirical Bayes linear regression model (Bioconductor package: Limma). This signature was mainly determined by overexpression of the BRE (brain and reproductive organ-expressed) gene. The mean average VSN normalized expression for BRE in the t(9;11)(p22;q23) subgroup was 3.7-fold higher compared with that in other MLL-rearranged cases (p<0.001). Validation by RQ-PCR confirmed this higher expression in t(9;11)(p22;q23) cases (p<0.001). In addition, we confirmed that overexpression of BRE was predominantly found in t(9;11)(p22;q23) in an independent gene expression profile cohort (Ross et al, Blood 2002). Remarkably, MLL-rearranged cases with a BRE expression higher than the mean expression showed a significant better 3 year disease free survival than MLL-rearranged cases with a lower expression (80±13% vs. 30±10%, p=0.02).

Previously, overexpression of BRE has been described in hepatocellular carcinomas (HCC) (Chang et al., Oncogene 2008) and an anti-apoptotic effect was described. We transfected BRE in the monomac-1 cell line, which harbors a t(9;11)(p22;q23). We did not find a proliferative advantage for BRE overexpression using a BrDU-assay nor changes in drug sensitivity, indicating that the anti-apoptotic effect as described for HCC in vivo could not be confirmed in vitro in AML.

In conclusion, overexpression of the BRE gene is predominantly involved in pediatric MLL-rearranged AML with t(9;11)(p22;q23). Moreover, high expression of BRE showed a favorable prognosis. We did not find any influence of BRE expression on cell proliferation or apoptosis in vitro. This indicates that further studies involving the role of the MLL-fusion protein on BRE transcription are necessary to unravel the leukemogenic role in pediatric AML.