Defining CP-CML patient subsets associated with poor imatinib uptake and response.

The introduction of tyrosine kinase inhibitor (TKI) therapy, specifically imatinib, has dramatically improved the treatment outcome for the majority of chronic phase chronic myeloid leukaemia (CP-CML) patients. Although most patients will achieve excellent clinical (haematological, cytogenetic and molecular) responses on imatinib, it is clear that a subset of patients will respond poorly, or fail imatinib therapy. Currently, up to 35% of patients treated with imatinib fit into this subset, displaying either primary or acquired resistance, leading to sub-optimal response or imatinib failure. The organic cation transport-1 (OCT-1) protein is the major active protein involved in imatinib transport. Measuring the function of OCT-1 in leukaemic mononuclear cells prior to imatinib therapy, expressed as OCT-1 activity (OA), has been demonstrated to be a strong prognostic indicator. Notably, low OA is strongly associated with patients at significant risk of poor molecular response, mutation development and leukaemic transformation during imatinib therapy. It is important to therefore determine what factors underlie the range of OA levels observed in CP-CML patients, and whether patients with very low OA and poor response to imatinib have different overall disease characteristics associated with alternative biological mechanisms. The present study sought to 1) determine the variation in CP-CML patient immunophenotype at diagnosis, in relation to patient characteristics, including OA; 2) determine the gene expression patterns associated with OA, and identify new biomarkers for CP-CML; and 3) determine the global DNA methylation profile of CP-CML, with particular focus on very low OA, and ascertain whether aberrant epigenetic programming may underlie poor imatinib response. Specific lineage differences were identified, with patients defined as very low OA associated with a decreased T-lymphocyte signature, compared to very high OA. Furthermore, an up-regulated histone gene signature associated with very low OA was identified. Gene expression analysis also identified GFI1 as a novel biomarker for progression in CP-CML, as patients with low diagnostic GFI1 expression in their white cells were at significant risk of disease transformation to blast crisis (BC), even when receiving TKI therapy. Additionally, significant differences in global DNA methylation patterns were identified between CP-CML and normal individuals; CP-CML patients with very low OA, compared to all other patients; and CP-CML versus BC. Importantly, this is the first report of global DNA methylation analysis in CML and identifies that aberrant epigenetic programming may have a significantly greater role in CML than originally first thought. In conclusion, the findings detailed in this thesis provide further insight into the heterogeneity of CP-CML, and to a lesser extent OA. Additionally, a greater understanding of the possible factors influencing OA determination is presented, along with a new biomarker (GFI1) for disease progression to blast crisis in de novo CP-CML. Finally, the global DNA methylation results may present novel targets and pathways responsible for poor TKI response that will assist in developing new therapeutic strategies for de novo CP-CML patients, involving combination therapy to enhance patient outcome.Thesis (Ph.D.) -- University of Adelaide, School of Medicine, 201