[CMR[B] Lecture] Predicting the unseen: Using mathematical models to speculate about biological mechanisms in hematopoiesis
Ingmar Glauche, PhD
Group Leader and Deputy Head of Institute, Institute for Medical Informatics and Biometry (IMB), Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden, Germany
Mathematical models can be used to describe biological processes on both a conceptual and a quantitative level. Their suitability is usually evaluated by the models’ ability to consistently mimic different critical phenomena. It is the challenge to derive models with a level of complexity that can be validated based on the available data and which provide additional, potentially testable predictions about unobserved quantities. Based on our current research I will present two examples in which we use mechanistic mathematical models to understand organizational principles in healthy and diseased hematopoiesis. The first approach focusses on temporal readouts of clonally marked, hematopoietic cells. We use different computational approaches to investigate how and under which conditions the mechanisms of self-maintenance and differentiation of the stem cell population can be reconstructed. The second approach addresses the treatment of chronic myeloid leukemia patients with tyrosine kinase inhibitors. Here, we use scale estimates based on clinical data to identify that the long-term response is not limited by the drug toxicity but by the rare activation of leukemic stem cells. We present strong evidence to suggest that TKI dose reduction should be considered as a long-term treatment option for CML patients and that this approach provides further predictions about the patient’s response after potential treatment cessation.