Modelling Cellular and Molecular Progression Of CML In The Mouse

Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm, caused by a reciprocal chromosomal translocation that generates the BCR-ABL fusion protein, a constitutively activated tyrosine kinase. Patients with CML usually present in an indolent chronic phase (CP), however, if left untreated, they irrevocably progress to an aggressive form of acute leukemia, termed blast crisis (BC) that is usually fatal. Tyrosine kinase inhibitor (TKI) (e.g. Imatinib) treatment has revolutionised the treatment of CML CP. However, ∼5-10% of CP patients will progress to BC despite TKI treatment, and an additional 10-15% of patients are beyond CP at initial presentation. Upon disease progression, treatment options are very limited and prognosis is dismal. Hence, understanding the events that drive disease progression and identifying potential therapeutic targets remains an unmet clinical need. The mechanisms of BC transformation are poorly understood, but it is generally accepted that additional somatic mutations are required. To date, a small number of recurrent mutations have been reported, but these only account for a relatively small number of cases and their exact nature is not fully understood.

In order to study the mechanisms of BC progression and to identify the co-operating mutations that drive this, we have utilised a published transgenic murine model of chronic phase CML (Koschmieder et al., 2005) and performed a transposon-based forward insertional mutagenesis study. In our mouse model, expression of BCR-ABL was driven in the hematopoietic stem and progenitor cell compartment (HSPC) by an SCL enhancer in a tetracycline dependant manner. Following BCR-ABL expression we conditionally induced ongoing mutations via a transposon-transposase system (SB) within HSPC and monitored disease progression from the chronic/BCR-ABL dependant phase to the transposon-mediated BC. Utilising the design of the transposon based system, it was then possible to identify these mutations by multiplexed next generation sequencing (NGS). Our experimental cohort was comprised of BC mice (which expressed BCR-ABL and transposon/transposase mediated mutation induction), CML mice (BCR-ABL only) and SB mice (mutation induction only). BC mice demonstrated a significantly shorter survival (p<0.0001, 116 vs. 147 days) compared to CML mice. Disease progression was characterised by a significantly increased disease burden, in terms of organ infiltration and leucocytosis, with around 80% of BC mice developing an exclusively acute myeloid leukemia by the Bethesda criteria. BC mice also demonstrated quantitative and functional differences within the hematopoietic stem and progenitor cell compartment in in vitro and in vivo assays in keeping with progression from a chronic to an acute leukemia. Importantly, BC mice showed a shorter survival (p=0.007, 116 vs. 128 days) compared to the SB mice, in which both acute myeloid and lymphoid leukemias were seen. Molecularly, NGS revealed insertions in both novel genes, and in genes previously implicated in CML blast crisis, hematopoiesis and leukemogenesis, such as ASXL1, FLT3 and ERG. These insertions included highly recurrent hits and were enriched for transcriptional regulators and signalling proteins, many of potential therapeutic relevance. Additionally, there was only a very modest overlap between the mutations identified in the BC and the SB cohorts, demonstrating BCR-ABL-dependant cooperation and disease progression. Considering the above data, our mouse model shows great potential in understanding the mechanisms of transformation to blast crisis, and ultimately in identifying potential therapeutic targets.