The application of proton computed tomography to reduce proton therapy range uncertainties

Proton therapy is a method of radiotherapy utilising protons to deliver a therapeutic dose of radiation to target cancer. Unlike x-rays, protons in the therapeutic energy range (< 250 MeV) have a finite range in the body. The physics of proton interactions mean that protons deposit most of their radiation dose at the end of their range. Hence, through careful selection of proton energy, protons have the potential to deliver dose to the target whilst sparing healthy surrounding tissue, as well as reducing the total dose given to the patient. This is particularly favourable for paediatric patients. However, the accuracy of proton therapy is currently limited by uncertainty in the delivered proton range. Because of this range uncertainty, a margin of typically 3.5 mm + 3% is added to the proton range. A major source of range uncertainty in proton therapy arises from the use of x-ray CT when imaging the patient for treatment planning. Here, an alternative imaging modality is tested in an effort to reduce range uncertainty. A proposed solution to remove this source of uncertainty is the use of proton CT. In proton CT, the stopping power relative to water (RSP) of the patient is measured directly, potentially increasing the accuracy of imaging for proton therapy treatment planning. The PRaVDA prototype proton CT system is a proton-tracking CT system designed using fully solid-state technology to resolve the paths of individual protons entering and exiting a phantom, and then measure the residual range of the phantom. With this information, an image is constructed of proton stopping power. In this thesis, the first results from the PRaVDA proton CT system are xiv shown. An image of a test phantom was acquired and the RSP accuracy of the image was shown to be better than 1.3% in materials replicating soft tissue and bone. The image contained artefacts arising from the raw data acquired and the source of these artefacts is investigated. Further study into the use of proton CT for proton range calculation was performed using a dosimetric phantom. The dosimetric phantom contains a section of EBT-3 radiochromic film capable of measuring a 2D dose distribution. The phantom was exposed to proton beams at two different energies, and images of the phantom were acquired using proton CT and x-ray CT. The proton CT and x-ray CT images were used to calculate the expected proton range using a validated Monte Carlo simulation, and the simulated results were compared against the experimental measurement.