Angiogenesis is a critical step in
tumor growth, development, and
invasion. Nascent tumor cells secrete vascular endothelial growth
factor (VEGF) that significantly remodels the tumor microenvironment
through interaction with multiple receptors on vascular endothelial
cells, including type 2 VEGF receptor (VEGFR2). The complex pathways
initiated by VEGF binding to VEGFR2 lead to enhanced proliferation,
survival, and motility of vascular endothelial cells and formation
of a new vascular network, enabling tumor growth. Antiangiogenic therapies
that inhibit VEGF signaling pathways were among the first drugs that
targeted stroma rather than tumor cells. Despite improvements in progression-free
survival and higher response rates relative to chemotherapy in some
types of solid tumors, the impact on overall survival (OS) has been
limited, with the majority of tumors eventually relapsing due to resistance
or activation of alternate angiogenic pathways. Here, we developed
a molecularly detailed computational model of endothelial cell signaling
and angiogenesis-driven tumor growth to investigate combination therapies
targeting different nodes of the endothelial VEGF/VEGFR2 signaling
pathway. Simulations predicted a strong threshold-like behavior in
extracellular signal-regulated kinases 1/2 (ERK1/2) activation relative
to phosphorylated VEGFR2 levels, as continuous inhibition of at least
95% of receptors was necessary to abrogate phosphorylated ERK1/2 (pERK1/2).
Combinations with mitogen-activated protein kinase/ERK kinase (MEK)
and spingosine-1-phosphate inhibitors were found to be effective in
overcoming the ERK1/2 activation threshold and abolishing activation
of the pathway. Modeling results also identified a mechanism of resistance
whereby tumor cells could reduce pERK1/2 sensitivity to inhibitors
of VEGFR2 by upregulation of Raf, MEK, and sphingosine kinase 1 (SphK1),
thus highlighting the need for deeper investigation of the dynamics
of the crosstalk between VEGFR2 and SphK1 pathways. Inhibition of
VEGFR2 phosphorylation was found to be more effective at blocking
protein kinase B, also known as AKT, activation; however, to effectively
abolish AKT activation, simulations identified Axl autophosphorylation
or the Src kinase domain as potent targets. Simulations also supported
activating cluster of differentiation 47 (CD47) on endothelial cells
as an effective combination partner with tyrosine kinase inhibitors
to inhibit angiogenesis signaling and tumor growth. Virtual patient
simulations supported the effectiveness of CD47 agonism in combination
with inhibitors of VEGFR2 and SphK1 pathways. Overall, the rule-based
system model developed here provides new insights, generates novel
hypothesis, and makes predictions regarding combinations that may
enhance the OS with currently approved antiangiogenic therapies.