Spatio-temporal mechanistic modeling of whole-cell tumor metabolism - Abstract
Understanding the metabolic characteristics of tumors and their environments is crucial for elucidating the
mechanisms of cancer development and for developing therapeutic strategies. Despite the increasing availability
of 3D gene expression and other high-throughput data, a major unresolved challenge is how to translate complex
datasets and knowledge of human metabolism and cellular biophysics into forecasts of tumor growth dynamics,
spatial structure and severity, and possible therapeutic strategies. Our highly interdisciplinary project will
leverage existing computational approaches to address this challenge, establishing a new avenue for performing
spatio-temporal modeling and simulations of whole-cell cancer metabolism in its microenvironment. Previous
work has explored 3D mathematical models of cancer growth based on simplified descriptions of cell populations,
e.g. through differential equations. In parallel, based on the approach of flux balance analysis, detailed tumor
metabolism models have been used to predict all steady state fluxes in the cell, and the effects of perturbations
of target genes. While in principle possible, models combining 3D spatio-temporal dynamics with detailed
genome-scale metabolism, have not been developed yet. Here, we propose to repurpose our free and open-
access software platform for computation of microbial ecosystems in time and space (COMETS) towards the
study of tumor growth dynamics. Specifically: Aim 1: We will generate omics-data-constrained genome scale
models of specific cancer cell lines, and import them into COMETS. We will then simulate overall tumor growth
dynamics, and test our capacity to accurately predict key metabolic phenotypes, such as growth curves, glucose
and amino acid uptake, and lactate secretion. Aim 2: We will build upon our capacity to accurately simulate with
COMETS fine details of multicellular dynamics in 2D to generate and test predictions of tumor growth on a
surface. We will vary tumor geometry and microenvironment composition, and experimentally test predictions
using a cancer on-chip approach. Aim 3: Using the advanced capabilities of COMETS, we will explore tumor
heterogeneity, and extend our detailed biophysical model for biomass propagation to 3D realistic
microenvironments (with gradients and vascularization), in search for metabolic characteristics associated with
morphological features of 3D tumors. We expect that results generated through this project will pave the way for
predictive modeling of cancer growth and metabolism, applicable to the study of in vivo tumors. Gradual
application of new COMETS capabilities will allow us to extend initial models to more complex scenarios and
configurations, including interactions between different cell types, detailed modeling of specific tumor geometries
based on imaging data, predictions of metastasis and metabolic adaptation in tissues other than the tissue of
origin, simulations of interactions with the microbiome, and the implementation of in silico testing of thousands
of combinatorial therapeutic strategies.
