An integrated molecular study of human acute myeloid leukemia (AML) with a specific focus on genome wide methylation in patients with aberrant EVI1 expression.

Acute myelogenous leukemia (AML) may appear as a homogenous disease, but the underlying pathology is highly heterogeneous, which has major consequences for the development of the disease and prognosis of survival. Well studied AML subtypes are those characterized by the expression of ‘novel’ transcription factors that are fusion proteins such as AML1-ETO, CBFbeta-MYH11 or PML-RAR fusion genes as the result of chromosomal translocations t(8;21), inv(16) or t(15;17) respectively. In many leukemias the genetic abnormalities are still unknown, whereas in others genetic lesions and corresponding genes have been identified, but the function of those genes and the apparent role in transformation remain frequently elusive. A particular example of such a leukemia type is AML carrying 3q26 lesions, associated with excessive EVI1 expression. Patients with this AML type are extremely difficult to treat and insight into the biology of the disease is warranted. Expression profiling and mutation analysis have given important new insights into the heterogeneity of AML, in some cases into underlying pathology, and into its relevance for therapy response. Although it has become evident that not only genetic changes but also epigenetic alterations may play an important role in tumorigenesis, very little is known about altered methylation of critical genes in human AML. In this proposal we wish to explore a novel array-based DNA methylation profiling assay (HELP) (1) in AML to identify cases with profoundly distinct methylation profiles and to discover novel disease genes whose expression has been altered epigenetically and that consequently play a critical role in transformation. We expect that a multidimensional classification of AML, taking (i) silencing by promoter methylation, (ii) gene expression (2-5) and (iii) loss of heterozygosity (high resolution SNP array analysis) (6) into account, will be instrumental to identify cooperating mechanisms in AML. Depending on the type of epigenetic or genetic alteration that will be found, we may develop assays to be used prospectively in new AML cases. Although development of these tools will be of major consequence for many leukaemia types, we will here particularly apply the tools to AML with 3q26 lesions. The main objective of this program is to unravel the epigenetic and genetic complexity of AML carrying 3q26 lesions and develop diagnostic tools that give insight into the pathological events leading to this disease type. We will 1) explore the methylation-profiling assay for human AML, 2) develop comprehensive software tools to analyze gene methylation profiling data in combination with gene expression profiles and LOH data obtained with high resolution SNP array analysis, and 3) We propose to apply these tools specifically to 3q26 AML and develop novel diagnostic tools based on our findings for this AML subtype. The tools will be developed such that also other AML subtypes may be studied in the future and by other scientists. 4) Based on these tools we anticipate on the identification of novel tumor-suppressor genes and also define unique previously unrecognized AML subtypes.