Modeling of radiosensitization effecct in photon irradiated cells with embedded high-z nanoparticles

Abstract

Radiotherapy is one of the most important methods of cancer treatment. Despite that, radiotherapy possesses one major limitation: ionizing radiation does not discriminate between normal tissue and malignant tissue, which results in high morbidity of healthy tissue near the tumor. Combining radiotherapy with high-Z nanoparticle (NP) radiosensitizers provides great opportunities for increasing treatment efficiency and widening the therapeutic window. In order to predict the radiosensitization effect of NPs, a modification of the local effect model, in which the energy deposition from NPs is assessed by Monte Carlo (MC) radiation transport codes, has been employed in the past. In this thesis, a combined framework that splits the consideration of the radiosensitization effect into two steps is developed in order to have an approach to the biological outcomes of the radiation treatment assisted with high-Z NPs. The first step is the evaluation of the radial dose distribution (RDD) around a single NP ionized by a photon beam with a given energy spectrum using MC simulation. MC simulations were performed using TOPAS (TOol for PArticle Simulation), which is based on Geant 4, a toolkit for the simulation of the passage of particles through matter. To simulate the RDD, single spherical gold NPs of various diameters suspended in water were irradiated with mono-energetic and poly-energetic photon beams, and produced secondary electrons were scored at the NP surface as phase space. After that, the scored electron spectra were used to evaluate the deposited doses in water-like tissue around irradiated NPs, which, in turn, were fitted by a power law function. In the second stage, an analytical approach based on the local effect model and the simulated RDDs was used to evaluate the average dose and the average number of lethal lesions in a cell target due to a set of irradiated NPs. The general expressions (in the integral form) for the average number of lesions and the survival probability of irradiated cells loaded with high-Z NPs were derived. For the case of a spherical cell target and the power law RDD the integrals were taken using the Maple package and the explicit analytical expressions describing the average doses and the average of the squared doses per one ionized NP located at a given distance from the target center were derived. The expressions contain only four parameters: the target (nucleus) radius and the other three related to the power law approximation of the NP RDDs, which are the NP radius, the dose per one ionization at the NP surface, and two parameters describing the power law RDD. Integration of these expressions over a cell volume with a given distribution of NPs allowed us to derive an expression from calculating the survival probability and related metrics quantifying the biological effects (relative biological effectiveness and other similar). The derived expressions were applied to calculate the survival curves and relative biological effectiveness for a culture of spherical cells loaded with gold NPs and irradiated with monoenergetic photons of 10 - 150 keV. The proposed framework provides a practical alternative to time-consuming MC simulations, enabling the assessment of the response of cell cultures to an irradiation treatment assisted with NPs for a wide variety of cell geometries, NP distributions, and irradiation schemes. The validation of the proposed model was performed, evaluating the dependence of RBE on the GNP radius and the variation of RBE with the photon energy. The results present an excellent agreement with results reported in the literature, allowing a successful approach to predicting the radiosensitization effect of irradiated high-Z NPs on human cancer cells.