Simulating Space Radiation-Induced Breast Tumor Incidence Using Automata.

Area of Science:

• Radiation biology
• Computational modeling
• Cancer epidemiology

Background:

• Estimating cancer risk from space radiation is difficult due to reliance on atomic bomb survivor data (acute gamma rays) versus chronic high-LET cosmic radiation.
• Existing models struggle to accurately predict radiation-induced cancer due to complexities of different radiation types and effects.

Purpose of the Study:

• To develop and validate a cellular automata model for estimating long-term cancer risk from different radiation qualities.
• To investigate the roles of targeted (TE) and non-targeted radiation effects (NTE) in radiation-induced breast cancer.
• To predict the relative biological effectiveness (RBE) for breast cancer induction by cosmic radiation.

Main Methods:

• Utilized a cellular automata-based two-stage clonal expansion model.
• Validated and tuned model parameters against spontaneous breast cancer incidence in an unexposed population.
• Incorporated both targeted effects (DNA damage, cell death) and non-targeted effects (genomic instability) into the model.
• Compared model predictions with epidemiological data from atomic bomb survivors.

Main Results:

• Targeted effects alone were insufficient to accurately predict radiation-induced cancer.
• Non-targeted effects, lasting approximately 100 days post-irradiation, were necessary for accurate dose-dependent predictions of gamma-ray-induced breast cancer.
• The model predicts a maximum RBE for breast cancer induction by cosmic radiation at 220 keV/μm when considering both TE and NTE.

Conclusions:

• A cellular automata model incorporating both targeted and non-targeted radiation effects can accurately predict breast cancer risk.
• Non-targeted effects play a critical role in long-term cancer risk assessment from radiation exposure.
• This modeling approach provides a foundation for studying chronic low-dose exposures and complex space radiation scenarios.