Modelling disease progression in myeloproliferative neoplasia patients

Myeloproliferative neoplasms (MPN), e.g. such as JAK2-mutated MPN and BCR-ABL-positive CML, represent very attractive diseases to study multi-step malignant transformation in vivo and in vitro for the following reasons: these diseases represent molecularly well-defined malignant stem cell disorders that follow a robust and characteristic clinical course. The malignant stem cell compartment seems rather oligoclonal at disease onset and can then be treated successfully with molecular targeted TKI-based approaches. Still, the different MPN subtypes are (although at different frequency and kinetics) typically characterized by eventual disease progression from a rather “benign” early chronic state both towards myelofibrosis and/or leukemic transformation. The identification of biomarkers that are able to reliably predict the individual transition from the chronic state (characterized by increased cellular turnover) to the more malignant, aggressive state (due to the acquisition of a additional differentiation block) of the disease is crucial for therapy, but extremely challenging because of the heterogeneity of the cell populations involved, the complexity of the underlying mechanisms as well as their mutual interactions. The extraction of predictive models out of complex multivariate data sets accessible today requires the development of advanced computational methodologies as well as adjusted experimental and clinical validation.

In chronic myeloid leukemia (CML), we have previously been successful in the modelling of predominant modes of action from data analyzing patterns as well as kinetics of resistance development to tyrosine kinase inhibitors (TKI). By working on the identification of biomarker signatures characteristic for the transition from chronic phase (CP) CML to accelerated phase (AP) or blast crisis (BC), we can show that the models fit well to the generic concepts of transition states in complex systems.

 

 

These findings indicate optionally generic applicability of these concepts which will now be extended to other types MPN by using available in silico data complemented by newly generated “-omics” data on well-defined subpopulations of patient cells. Results will be validated clinically within the SAL-MPN registry as well as clinical prospective trials.

Selected references:

 

People