Computational signaling models to fight Tyrosine Kinase Inhibitors resistance in leukemias

Intracellular signaling dysregulation is a key factor in many pathological processes, including the onset of resistance to anti-cancer therapies. In both acute and chronic myeloid leukemias, aberrant activation of protein tyrosine kinases (PTKs), such as FLT3 and BCR-ABL, sustain the proliferation of leukemic cells. While Tyrosine Kinase Inhibitors (TKIs) have been developed to target these oncogenic receptors, 20-30% of patients develop resistance, often due to the activation of alternative signaling pathways. In this doctoral thesis, I employed two different yet complementary computational approaches to unravel TKI resistance mechanisms in AML and CML and identify combination therapies to restore treatment sensitivity. First, I developed SignalingProfiler, a novel computational strategy that bridges multi-omics data to cellular phenotypes using literature-derived causal signaling networks. This approach allowed for the construction of context-specific models of intracellular signaling, unveiling new resistance axes in AML and CML and highlighting promising drug candidates for combination with TKIs. Second, I generated genotype-specific Boolean models of FLT3-ITD positive AML patients. These models were used to screen different drug combinations in silico and represented a step toward the usage of computational models within a translational context. Together, these strategies not only have deepened our understanding of the molecular mechanisms driving TKI resistance, but also paved the way for the development of more effective, targeted, and personalized therapies for acute and chronic myeloid leukemia patients.