Cancer Research
cancerresearch2
TUMOR MODELS

 

 

Computer simulation of tumor growth (vasculature and signaling molecules)

Computer simulation of tumor growth (vasculature and signaling molecules).

 

 

 

Tumor Models

Background

Devising cancer treatments requires an extensive understanding of tumor biology, including tumor formation, growth, and spreading; interactions between malignancies and their microenvironment or the surrounding tissues; and the role of vasculature. At IT'IS we study these dynamical processes from the chemical to the macroscopic level using in silico models of tumors. These highly flexible and pliable models can help us understand the physical and chemical processes occurring within tumors at various stages of cancer progression, the mechanisms of growth and angiogenesis, and the detailed temperature-induced effects during thermal cancer treatment for instance. They will become increasingly realistic and useful as more biological and physiological information is integrated.

Selected Past Achievements

  • Development of multiphysics, multiscale models that couple (non-linear) mechanical effects (e.g., from cell proliferation, apoptosis and migration); and explicit and implicit perfusion models with multiple convection-reaction-diffusion equations. Variables include different cells (healthy cells; proliferating, quiescent and necrotic tumor cells; endothelial cells), oxygen and other nutrients, signaling molecules and external factors.
  • Reproduction of typical tumor features in the models (e.g., capsule with brush border and necrotic core, exophytic tumor morphology, and vasculature development)
  • Investigation of the impact of various treatments (e.g., drugs, such as chemotherapy drugs or vasculature disruptive agents, and heat-based treatments) on tumor morphology, vasculature, and growth.

Next Challenges

  • To extend the models, e.g., to consider additional factors such as interstitial fluid pressure
  • To identify and perform suitable biological experiments to validate prediction models.