Degree Name

Doctor of Philosophy


School of Physics


Particle therapy is a radiotherapy modality in which accelerated protons or heavier ions are used to deliver a therapeutic dose to a treatment region. There are several advantages to using particle therapy over conventional photon therapy, including higher dose conformity, a higher relative biological effectiveness and increased healthy tissue sparing (particularly in the region behind the target). Due to the high dose conformity, large dose gradients will be present at the boundary between the treatment region and surrounding healthy tissue. Therefore, improvements in quality assurance techniques and development of methods to further improve the target-selectivity of particle therapy are two of the most important opportunities for enhancing the delivery of particle therapy.

During particle therapy, nuclei within the beam or target may fragment as the beam travels through a patient, resulting in the production of a range of secondary fragments. Some of these fragments are unstable nuclei, including several which decay via positron emission. Imaging of positron-emitting fragments through positron emission tomography (PET) has been proposed as a form of quality assurance in particle therapy. These may be compared with Monte Carlo simulations of the planned treatment for indirect verification of dose delivery, or used as input for algorithms which directly estimate the delivered dose based on the positron-emitting fragment distribution. An essential part of the development of these dose quantification algorithms is extensive Monte Carlo simulations in a variety of homogeneous and heterogeneous materials. If the simulation results are to be successfully translated to clinical practice, it is critically important that the models used to predict the production and distribution of positron-emitting fragments are accurate, consistent and experimentally validated. However, due to the broad range of energies and variety of different nuclear interactions, the modelling of the fragmentation process is extremely complex, and many different simulation models are available. More accurate estimation of dose distribution may also be achieved if the signal to noise ratio of the PET images can be improved.

An additional research challenge is improving the target-selectivity of particle therapy, potentially enabling further improvements in tumour ablation and tissue sparing. Tumour radiosensitisation methods have been widely investigated in photon therapy, but to date, there has been little research on the potential for radiosensitisation in particle therapy. The unique characteristics of the radiation field inside the treatment volume present new opportunities for radiosensitisation, which have yet to be extensively investigated.

In this Thesis, improvements to dose quantification and new radiosensitisation techniques for particle therapy are investigated.

The accuracy of different hadronic inelastic fragmentation physics models available in different versions of the Geant4 Monte Carlo toolkit are compared to experimental measurements to determine which physics model most accurately predicts the spatiotemporal distribution of positron-emitting fragmentation products generated during carbon and oxygen ion therapy. The most accurate fragmentation physics model was found to be the Binary Ion Cascade model implemented in Geant4 version 10.2.p03.

It has been proposed that radioactive positron-emitting ion beams can be used for particle therapy, with the aim of improving the signal to noise ratio of quality assurance PET images. However, before such radioactive beams can be adopted therapeutically, it is necessary to both quantify their biological impact in comparison to stable ion beams, and quantitatively evaluate the achievable improvements in image quality which they will offer. In this Thesis, the relative biological effectiveness and PET imaging characteristics of radioactive primary beams are investigated using both Monte Carlo simulations and experimental measurements and compared to those of their respective stable isotopes. It was found that radioactive primary beams significantly increase the quality of PET images for the same given imaging time, while the relative biological effectiveness of the radioactive and stable beams are essentially indistinguishable.

Finally, this Thesis proposes and evaluates a novel radiosensitisation technique which utilises tumour-specific neutron capture agents to increase the dose to the tumour during particle therapy, by exploiting the thermal neutron field which exists in and around the target volume. The feasibility of this technique, called neutron capture enhanced particle therapy (NCEPT), is first established via Monte Carlo simulations. The magnitude and distribution of the thermal neutron field present during particle therapy is experimentally evaluated. Finally, the impact of using neutron capture agents based on boron-10 and gadolinium-157 isotopes during carbon and helium ion therapy on cultured human glioblastoma cells is experimentally evaluated. It was found that the addition of neutron capture agents to enhance the dose during particle therapy is feasible and effective, and results in a significant reduction in cell viability relative to unenhanced particle therapy.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.