Functionalist Oncology to Model the Contextuality of Dynamics and Treatment in Acute Myeloid Leukaemia

Abstract

Acute myeloid leukaemia (AML) is a disease with high intra- and interpatient heterogeneity. Current treatments are still based on cytarabine combined with an anthracycline, with 5-year overall survival rates below 30%. Functionalist oncology utilises edge-weighted digraphs as a representational tool to model the mathematical functional interdependencies of disease factors. This methodology enables a conceptual and mechanistic analysis of key factors influencing therapeutic success, providing a framework that can be parameterised to explore patient-specific treatment responses. It incorporates the oligoclonal nature of AML to perform, in principle, comprehensive cost-benefit analyses regarding the addition of individual drugs to existing treatment regimens and the number and type of chemotherapy courses to be given. Extrapolating from experimental data, we investigate the therapeutic potential and risks of SAMHD1 inhibitors in AML treatment as a proof-of-concept. We also provide a web-based interactive application to visualise hypothetical AML treatment scenarios, which can be combined with ex vivo single-cell phenotypic (gene expression), genotypic (somatic mutations) and functional (drug responses) analyses. Functionalist oncology can thus be used to generate testable hypotheses that might contribute to improving oncological decision-making, for example, by identifying the optimal number, nature and sequence of chemotherapy courses, including both existing and novel drugs, such as SAMHD1 inhibitors.