Therefore, the pathogenesis of HFRS is not associated with a direct virus effect

K/STAT signaling pathway were enriched. 9 Oncogenic Potential of FLT3 Mutations To verify our in vitro data showing differences in STAT5 signaling regarding FLT3 mutation type we used predefined gene sets implemented in GSEA. Since NPM1 is correlated with DMXB-A FLT3-ITD and has significant impact on gene expression, we tried to avoid this confounding variable by analyzing different subgroups regarding NPM1 mutation status. We were able to demonstrate that predefined STAT5A_02 was significantly enriched in the NPM1-WT subgroup, whereas STAT5A_01 and STAT5B_01 were also significant in the NPM1-mutated subgroup comparing FLT3-ITD with FLT3-WT. FLT3-TKD in comparison to FLT3-WT showed significant enrichment of STAT5A_01 04 and 10 Oncogenic Potential of FLT3 Mutations Enrichment of predefined STAT5 target genes in FLT3-ITD STAT5A_01 No selection regarding FDR 29%, p = 0.07 NPM1 mutation FLT3ITD+ vs. FLT3WT Subgroup A NPM1 mutation + FLT3-ITD+ vs. FLT3-WT Subgroup B NPM1 mutation 2 FLT3-ITD+ vs. FLT3-WT FDR 27%, p = 0.03 STAT5A_02 FDR 9%, p = 0.02 STAT5A_03 n.s. STAT5A_04 n.s. STAT5B_01 FDR 9%, p = 0.007 FDR 26%, p = 0.05 n.s. n.s. FDR 23%, p = 0.008 n.s. FDR PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19630074 10%, p = 0.04 n.s. n.s. n.s. Enrichment of predefined STAT5 target genes in FLT3-TKD STAT5A_01 No selection regarding FDR 32%, p = 0.07 NPM1 mutation FLT3TKD+ vs. FLT3WT Subgroup A NPM1 mutation + FLT3-TKD+ vs. FLT3-WT n.s. STAT5A_02 n.s. STAT5A_03 FDR 17%, p = 0.01 STAT5A_04 FDR 30%, p = 0.08 STAT5B_01 FDR15%, p = 0.007 n.s. n.s. n.s. n.s. Subgroup B NPM1 FDR 5%, p,0.001 mutation 2 FLT3-TKD2 vs. FLT3-WT FDR 11%, p = 0.04 FDR 11%, p = 0.04 FDR 3%, p = 0.001 FDR 8%, p,0.001 Analysis of STAT5 target genes in context of FLT3-ITD and FLT3-TKD mutation status in primary AML bone marrow samples. To avoid the interference of NPM1 mutations subgroups were composed, according to NPM1 mutation status. The differences in transcription factor target gene sets result from different targeted motifs representing the broad spectrum of potential binding sites. FLT3-ITD was associated in all subgroups with significant enrichment of STAT5A_02 target genes and in no subgroup with enrichment of STAT5A_03 and STAT5A_04 target genes. FLT3-TKD showed a PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19632393 more diverse picture with enrichment of STAT5A_03 and STAT5A_04 target genes in some groups. In FLT3-TKD and NPM1-mutated patients no enrichment of STAT5 target genes could be detected. indicates the presence of a mutation; indicated the absence of a mutation; FDR: false discovery rate; p: nominal p-value; n.s.: not significant. doi:10.1371/journal.pone.0089560.t002 STAT5B_01 in the NPM1-WT subgroup. This data has to be interpreted with caution but suggests that depending on the type of FLT3 mutations different transcription factors are activated. Discussion The most common FLT3 mutations are ITD and mutations in the second TKD, which are frequently found in AML but also to a lower frequency in ALL. Other groups have shown the adverse impact of FLT3-ITD mutations on OS and RFS in the group of CN-AML which could be confirmed in our study. However, the results for FLT3-TKD mutations are inconclusive. Our group found a trend towards a better RFS in FLT3-TKD positive patients compared to FLT3-WT cases. However, this was not statistically significant. The FLT3-TKD mutations did not have a significant impact on OS or RFS in multivariable Cox regression analysis. The majority of FLT3-TKD mutations were lost at relapse, whereas no additional TKD mutations were detectable at re