Gene Expression Studies of Hematologic Malignacies

The general aim of this thesis was to characterize hematologic malignancies using gene expression profiling in order to obtain an improved classification and to study deregulated transcriptional networks in leukemia. In the first study (Article I), the gene expression profiles of hematologic cell lines were analyzed to investigate whether such cell lines maintain expression profiles that are characteristic of their primary genetic changes. By analyzing a large number of cell lines and primary pediatric leukemias, it was shown that immortalized hematopoietic cell lines with specific genetic changes maintain a characteristic gene expression pattern despite their diverse origin and numerous passages in vitro. In the second study (Article II), the gene expression patterns of a series of pediatric acute leukemias and normal hematopoietic cell subpopulations were ascertained, showing that pediatric leukemias segregate according to their lineages and primary genetic changes. For the first time, expression patterns of genes associated with the primary genetic changes were investigated in normal hematopoietic cells of different lineages and maturations, identifying genes that were preferentially expressed by the malignant cells only. This suggests that leukemic cells ectopically activate genes, which reflect regulatory networks of pathogenetic importance that also may provide attractive targets for future directed therapies. In the third study (Article III), the aim was to build gene expression classifiers that could predict clinically important variables. The k-nearest-neighbours algorithm was utilized to build gene expression predictors that could classify the acute leukemias according to lineage and genetic subtype with high accuracy. In addition, minimal residual disease status (MRD) in T-cell acute lymphoblastic leukemias (ALLs) with a high (>0.1%) MRD at day 29 could be predicted already at the time of diagnosis. In pediatric leukemias with uncharacteristic cytogenetic aberrations or a normal karyotype, two novel subgroups were identified. In the fourth and final study (Article IV), breakpoint mapping using fluorescence in situ hybridization (FISH) and gene expression analyses revealed that CCND2 was juxtaposed to the vicinity of the TCR alpha/delta locus, in t(12;14)-positive T-ALLs, resulting in aberrant expression. The t(12;14) represents the first neoplasia-associated translocation shown to result in overexpression of CCND2