Abstract
A novel and versatile “bottom-up” approach is developed to estimate the radiobiological effect of clinic radiotherapy. The model consists of multi-scale Monte Carlo simulations from organ to cell levels. At cellular level, accumulated damages are computed using a spectrum-based accumulation algorithm and predefined cellular damage database. The damage repair mechanism is modeled by an expanded reaction-rate two-lesion kinetic model, which were calibrated through replicating a radiobiological experiment. Multi-scale modeling is then performed on a lung cancer patient under conventional fractionated irradiation. The cell killing effects of two representative voxels (isocenter and peripheral voxel of the tumor) are computed and compared. At microscopic level, the nucleus dose and damage yields vary among all nucleuses within the voxels. Slightly larger percentage of cDSB yield is observed for the peripheral voxel (55.0%) compared to the isocenter one (52.5%). For isocenter voxel, survival fraction increase monotonically at reduced oxygen environment. Under an extreme anoxic condition (0.001%), survival fraction is calculated to be 80% and the hypoxia reduction factor reaches a maximum value of 2.24. In conclusion, with biological-related variations, the proposed multi-scale approach is more versatile than the existing approaches for evaluating personalized radiobiological effects in radiotherapy.
Highlights
Current biological models, including the commonly used linear-quadratic (LQ) model, are mostly “top-down” models[9]
Similar to the threshold and probability models used in track structure simulation, Monte Carlo damage simulation (MCDS) used four adjustable parameters and the optimal parameter values were determined from an empirical fit to the results from track structure simulations and radiobiological experiments
Total double-strand breaks (DSBs) yield is nearly constant at the energy level above 100 keV, whereas below 100 keV, it increases dramatically as incident energy decreases
Summary
Current biological models, including the commonly used linear-quadratic (LQ) model, are mostly “top-down” models[9]. The application of MC techniques for radiotherapy outcome prediction was highlighted by Naqa et al.[10] In their views, the multi-scale approach would encompass the multi-scale parts of the tumorigenesis (atomic, molecular, tissue, organ) and various multi-scaled radiation-induced response stages (i.e. physical, chemical and biological) over the spatial and temporal axes. The step-by-step tracking algorithm is extremely time consuming and have limited applications in higher dimensions[11,12,13] To overcome this disadvantage, a fast and quasi-phenomenological Monte Carlo damage simulation (MCDS) algorithm was developed by Semenenko et al.[14,15] to estimate the radiation-induced damages at cellular level. As organism cells have multiple damage repair mechanisms to ensure genomic integrity, it is important to quantify the effect in the radiobiological modeling. One advantage of TLK model is that it provides a satisfactory formalism in correlating the process of DSBs with cell killing[24]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
