Click here to


Are you sure ?

Yes, do it No, cancel

An Automaton Model of Tumor Growth with Response to Irradiation

L Wu1*, Y Watanabe2 , (1) University of Minnesota,Minneapolis, MN, (2) University of Minnesota, Minneapolis, MN


(Tuesday, 7/31/2018) 9:30 AM - 10:00 AM

Room: Exhibit Hall | Forum 4

Purpose: To introduce a novel automaton model of tumor growth in response to irradiation in both two- and three- dimensions.

Methods: To model the tumor growth, we considered three properties: the limitation of the number of cells per unit area/volume, the local condition of oxygen supply (or hypoxia), and the effect of stress/immunity on growth. The death rate of cancer cells depended on the cytotoxic behavior of cells in reaction to the immune system. The probability of cell death after irradiation depended on the oxygen content and the dose. When the cells were exposed to ionizing radiation, the oxygen permeability of the blood vessel changed as a function of the dose. Consequently, the oxygen distribution changed, resulting in a change in the growth rate after irradiation. In the simulation model, most parameters had its physical and biological meanings and could be estimated using published data. For validation, we compared the simulations to in vivo data obtained from the rat rhabdomyosarcoma tumor experiments and clinical data of metastatic brain tumors treated with Gamma Knife radiosurgery.

Results: We found that the model could replicate the experimental data by selecting appropriate model parameters. The influences of alpha and beta, cell cytotoxic rate, the oxygen supply distribution, and other relevant factors affecting tumor growth and its response to irradiation were studied. For example, we found that increasing the oxygen density made the radiation less effective for 20 Gy irradiation, but the oxygen density did not affect the radiation response for 30 Gy.

Conclusion: The proposed model included several biological processes, which were not considered within one model previously. The main challenge was the difficulty estimating proper values of the model parameters. However, once those parameters were determined, the model could reproduce the experimental results reasonably well.


Radiation Effects, Blood Vessels, Simulation


TH- Dataset analysis/biomathematics: Informatics

Contact Email