The research presented in this thesis investigates a method of objectively quantifying tissue stiffness post radiotheiapy. The project has involved the design, development and modification of an indentation system that was compared with the currently used subjective clinical assessment. The breast has been chosen as the prime site for investigation in the light of the high value placed on cosmesis by patients, the ease of access to these tissues, and the availability of the untreated breast to serve as a control.

In the various brachytherapy techniques the intent is to deliver as high a tumour dose as possible, limited only by surrounding normal tissue tolerance. The main feature of the techniques is very steep dose gradients, representing a potential limiting factor in accurate dose distribution measurements around sources. Dose distributions at distances less than 1 cm are therefore normally generated using either validated Monte Carlo (MC) simulations or standard dose calculation formalisms, for example that of AAPM TG 43, while dose measurements can only be performed at larger distances, normally greater than 1 cm from the outer dimensions of the source encapsulation Ge-doped silica fibres are a viable thermoluminescent dosimetry (TLD) system, providing good spatial resolution of approximately 120 urn, sensitive response to ionizing radiations, large dynamic dose range, good reproducibility and reusability, dose rate independence, minimal fading, resistance to water and low cost. Dosimetric characterisation of commercial Ge-doped silica fibres have been obtained by subjecting them to kilovoltage therapeutic x-ray radiation beams, verifying their use for brachytherapy sources; dose response, reproducibility and fading at 90 kVp and 300 kVp have been investigated. Central-axis depth doses have been obtained at the two accelerating potentials using different field size applicators, measurements being made using the fibres in both water and a GAMMEX RMr 4571 solid water phantom. Comparison has been made with central-axis depth doses, measured using a 0.6 cm3 graphite-walled ionisation chamber data and British Journal of Radiology Supplement 25 tabulated values (both in water). Ge-doped optical fibre dosimeters show good dosimetric response for low photon energies. These desirable characteristics support the use of these TL fibres as dosimeters for brachytherapy applications. Ge-doped optical fibre TL dosimeters have been used to measure the dose distribution around two Low Dose Rate (LDR) 125r seeds; model 6711, the new thinner model 9011 and a High Dose Rate (HDR) 192rr (MicroSelectron V2) source at proximal distances down to 1 mm, measured in a Perspex medium. The anisotropy has also been measured in Perspex, for distances from 10 to 100 mm from a LDR 1251 seed model 6711 centre, in 10 mm increments and at angles 10° to 90° in 10° increments from the seed central axis. Measured doses have been compared with calculations and treatment planning system (TPS) predicted doses for the same locations. Monte Carlo simulations were obtained using the EGSnrc \ DOSRZnrc codes and TPS predicted doses were obtained using the system VariSeed V8.0.2. For 1251 seed model 6711, the measurements agree with simulations to within 2.3 % ± 0.3 % along the transverse and perpendicular axes and within 3.0 % ± 0.5 % for measurements investigating anisotropy in angular dose distribution. Measured and Veriseed™ brachytherapy treatment planning system (TPS) values agreed to within 2.7 % ± 0.5 %. For the new thinner 1251 model 9011, dose measurements were in good agreement with simulations to within 2.1 ± 0.2 %, while dose measurements and doses obtained through use of the Variseed TPS agreed well, to within 2.2 ± 0.5 %. The above work has therefore demonstrated the applicability of Ge-doped optical fibres for use in brachytherapy.

The latter two decades of the last century saw significant improvements in External Beam Radiotherapy (EBRT), moved primarily by the advances in imaging modalities and computer-based treatment planning. These advances led to introducing the addition of a fourth dimension, time, to the three-dimensional EBRT arena. This new era in EBRT brings with it challenges and opportunities, in particular to compensate for the effect of respiratory-induced target motion and enhancing treatment delivery. Thus, characterising and modelling respiratory motion is of major importance in this research area. This thesis aims to enhance the understanding and control the effect of respiratory motion. As part of this work, the first principal component analysis (PCA) of respiratory motion is presented, as a basis for compactly and visually representing respiratory style and variation. These studies can be divided into two main aspects: firstly, understanding and characterising respiratory motion as the basis of any further steps towards compensating respiratory motion and secondly, utilising this knowledge in predicting and correlating internal and external respiratory motion in the abdominal thoracic region. This work has been developed starting with a piecewise sinusoidal model in an Eigenspace for modelling, Adaptive kernel density estimation (AKDE) for prediction and finally Canonical Correlation Analysis (CCA) for external-internal target correlation. A comparative study between these proposed approaches and state-of-the-art prior works showed promising results in terms of accuracy and computational efficiency: 20% error reduction compared to support vector regression (SVR) and kernel density estimation (KDE) and a significant reduction in computation speed during training stage. This journey into modelling and predicting respiratory behaviour has naturally raised questions of how best to track external motion. The need to track the surface with more than one marker, established within the aforementioned PCA analysis, motivates the desire for markerless tracking. Therefore, two different markerless systems have been studied, as potential solutions for this area, combined with a mesh model of the anterior surface. This suggests that the Microsoft Kinect camera is a promising low-cost technology for makerless respiratory tracking with less than 3.1 ± 0.6 mm accuracy.

The fluence correction factor, which accounts for the loss of primary protons and the production of secondary particles due to different non-elastic nuclear interactions at water equivalent depths in different phantom materials compared to water is an important parameter for the dose conversion in clinical proton dosimetry between any phantom material and water. Non-elastic nuclear interactions introduce uncertainties in the standard absorbed dose-to- water in radiotherapy. This thesis is part of an ongoing project at the UK National Physical Laboratory. (NPL) focussed on the development of graphite calorimeters for proton dosimetry. The fluence correction factor was investigated to give accurate dose conversions from dose- to-graphite in a graphite phantom to dose-to-water in a water phantom. The fluence correction factors at water equivalent depths have been studied for various dosimetric materials including A-ISO tissue equivalent, polymethyl methacrylate (PMMA), aluminium and copper with respect to water and with respect to graphite. The water equivalence of materials such as Plastic Water (PW), Plastic Water Diagnostic Therapy (PWDT) and solid water (WTl) phantoms was evaluated using a 60 MeV proton beam at the Clatterbridge Centre for Oncology. Plastic-water phantoms are widely used in radiotherapy as a substitute for water, in particular for non-reference dosimetry. However, while they are usually made 'water equivalent' for a particular beam type, they are not universally water equivalent due to their different elemental composition and associated different proton interaction cross sections (compared to water). Numerous studies of the water equivalence of plastic-water phantoms have been reported for photon and electron beams, but none with clinical proton beams. In the latter, non-elastic nuclear interactions take place which could potentially influence the water equivalence. This thesis evaluates the fluence correction factor at equivalent depths for proton energies of 60 Me V and 200 Me V, with respect to both water and graphite. This work was performed using analytical model calculations (which incorporate the ICRU-49 (1993) stopping power data tables and ICRU-63 (2000) for the total nuclear interaction cross sections); Monte Carlo simulations using the FLUKA 2008.3 code; and also experimental work at the Clatterbridge Centre for Oncology (CCO) 60 Me V with both modulated and un- modulated proton beams. The analytical calculations for primary protons indicate an increase in the fluence correction at both low and high energies compared to the Monte Carlo simulations. When the secondary charged particle were considered in the calculation, the fluence correction factor with respect to water was in general close to the unity for graphite, A-I50, PMMA, aluminium and plastic-water materials in 60 MeV mono-energetic beam. For proton energies of 200 Me V, the fluence correction was found to increase to a the order of a IV Fluence correction (actor (or various materails in clinical poroton dosimetrv Abstract few percent. The experimental finding for modulated and un-modulated 60 MeV protons showed that the fluence correction factor with respect to water is close to unity for graphite and PWDT with an uncertainty of 0.2% at la. The derived fluence correction with respect to graphite was also close to the unity for A-ISO and plastic-water materials, however, it was found to increase with depth to approximately 4% and 6% for aluminium and copper respectively (in modulated beam). In general, the experimental results for modulated and un- modulated 60 MeV proton beams show good agreement with the Monte Carlo simulations for modulated and un-modulated beams, yielding small fluence corrections, within the statistical uncertainty. For 200 MeV protons, the Monte Carlo simulations showed that the correction with respect to water increased with penetration depth giving values of up to 4% for graphite and 1.5% for A-ISO, PMMA, aluminium, copper and plastic-water materials. The fluence correction with respect to graphite was found to vary with penetration depth and hence it can be concluded that fluence correction factors need to be applied to ensure accurate dosimetry for all of the materials used in the current work with a 200 MeV proton beam.

Low linear energy transfer (LET) ionizing radiation as gamma and X-rays is widely used in the modem medicine for diagnostic and treatment purposes. Despite all the advantages that it gives for the early diagnosis and in the radiotherapy treatment of cancer, there are still unclear aspects about the mechanisms of the radiation effects in human tissue. Of particular interest is the detailed pathway which the directly irradiated cells use to communicate to their neighbours and its possible implications for radiotherapy applications. The aim of this project is to study the spatio-temporal signalling from irradiated to non- irradiated cells using in vitro 3D tissue models. For evaluation of radiation induced effects we used conventional 225 kVp X-ray and lead shielding to observe the effects in the non-targeted cells. Additionally we applied the novel 30 kVp micro collimator that could irradiate samples in wide 1-10 urn lines. The application of this targeted radiation source together with the shielding irradiation set up will cast light on the effects induced by localized radiation exposures and the signalling from the irradiated to non-irradiated cells using 3D organotypic skin as a model. Due to its fast cell turnover, the human epidermis is extremely sensitive to IR. This has a limiting effect on the radiotherapy as severe normal skin responses can delay treatment and even decline patients from radiotherapy. The current work is aiming to reveal the mechanisms of DNA damage, repair and their later consequences for the differentiation and development of inflammatory-like response in the irradiated and surrounding areas within the 3D organotypic skin model. We investigated use of inhibitors of pro-inflammatory pathways such as the transcription factor NF-KB and its downstream target COX-2, in order to reduce the signal transduction from the irradiated to non irradiared cells and to reduce the inflammatory responses in the surrounding normal tissue. Similar methods of control of the signal diffusion could be applied in the radiotherapy to reduce the inflammatory skin responses and decrease the range of the late side effects developing from the chronic inflammation that could further lead to ulceration and skin necrosis.

A 3D non-contact real-time multi sensor fringe measurement system is proposed in this research. The system is designed to help radiotherapists detect human body measurements and motions during the cancer radiotherapy treatment process in real-time. The ability to detect body movements will help radiotherapists to accurately estimate the actual radiation dose that was delivered to the tumour volume. The researchers at JMU built a three-sensor system that covers 3600 around the object so that the system can have full visibility around the human body under treatment. Each sensor consists of one camera and one projector and each sensor works on its own exclusive light bandwidth to enable all sensors to work concurrently without affecting each other's operation. Each sensor is connected to its own data processing unit that controls its operation and carries out data processing tasks and all three data processing units are connected to a main central computer that controls and coordinates the operation of the three sensors. The author developed a novel empirical calibration approach to calibrate the three sensors concurrently; the proposed calibration approach solves the depth calibration problem and the perspective problem to provide the system with the ability to measure 3D shapes in an absolute physical coordinate system that is consistent with the coordinate system of the radiotherapy treatment machine. The author also implemented two algorithms for absolute 3D body shape measurements. The first is a multi-frame algorithm using the phase-stepping technique while the second is a single-frame algorithm that is based on the Fourier Fringe Analysis technique. A performance comparison between these two methods is given in the literature. Finally, the author describes measurement validation and error reporting methods to provide the system with the ability to estimate the quality of its own measurements, to detect errors and report failures in the measurement process. This feature is very crucial to system operation especially in sensitive applications like cancer treatment as it provides qualitative and quantitative measures of validity to the measurements. This research has resulted in a practical real-time system that is implemented at both, the LAB and Christie Hospital Manchester. The system has been thoroughly tested and it has shown very good performance in terms of reliability, accuracy, usability and speed.

The human transferrin receptor (CD71) is a ubiquitously expressed internalising glycoprotein involved in iron homeostasis. Expression of the transferrin receptor is found to be greatly upregulated in tumour cells and, as such, it offers a useful target in the development of immunotherapeutic agents for tumour therapies. This laboratory has developed a Histagged anti-CD71 single-chain variable fragment (scFv) antibody, expressed in Pichia postoris, to be used in the investigation oftargeted radio-immunoconjugates against cancer. The production of an anti-CD71-scFv was confirmed using N-terminal sequencing, and the scFv was purified via a double His-tag using affinity chromatography to produce yields of 13.3mgfl of anti-C071-scFv. Specific binding of the scFv to the human transferrin receptor was evaluated by flow cytometry using C071+ HL-60 leukaemia cells and CHO cells that are CD7r. The anti-CD71-scFv was demonstrated to bind specifically to the CD71 TfR receptor with comparable binding to the monoclonal anti-C071 antibody using flow cytometry. No binding of either antibody to the COJ1- CHO cell line was observed. Saporin, a ribosome inactivating protein, was used to· demonstrate the internalising ability ofthe anti-C071-scFv. Saporin is highly cytotoxic but only if it can reach the cytosol of the target cell. The antiCD71- scFv was able to deliver a non-internalising monoclonal antibody-saporin conjugate intracellularly. This resulted in significant toxicity to CD71+ MCF-7 cells and confirmed that the anti-CD71-scFv could internalise.The anti-CD71-scFv was successfully conjugated to the y and Auger emitting isotope, 1:2.51. 12S1 was selected because the v-rays allow simple monitoring of the binding of the anti-CD71-scFvNal2S1 whilst the Auger electrons provide a potential cytotoxic effect if the radionuclide can get close to the nucleus. The anti-C071- scFv_Nal25 , was shown to retain specific binding ability to its target CD71 receptor, however Clonogenic and Comet assay systems did not show signs of DNA damage and cell cytotoxicity as a result of targeted radio-immunotherapy with the scFv-Naus, conjugate. The anti-CD71-scFv was shown to provide a useful delivery mechanism that is cheap to produce and purify and able to specifically bind and internalise into CD71+ tumour cell s. The isotope 12S1 was unable to demonstrate a cytotoxic effect on targeted cells, most likely due to the formation of monoiodotyrosine residues caused by the degradation of the anti-CD71-scFv intracellularly rapidly diffusing from the lysosome. Further investigation using alternative isotopes and the addition of nuclear localisation and endosomal escape sequences is warranted.

No description available.

High resolution voxel-based phantoms were used to assess organ doses from neutron external exposure. The segmented images of the Zubal voxel-based phantoms were obtained from the Imaging Science Research Laboratories at Yale University School of Medicine. The phantoms data were incorporated into MCNP4C2 and MCNPX Monte Carlo codes. There are twofold dose calculations in this study using monodirectional monoenergetic neutron beams in the energy range of 10 -9 (thermal) to 20 MeV, plus an extra calculation using a Maxwell fission spectrum source, under three different source irradiation configurations: anterior-posterior, posterior-anterior, and left lateral. Comparisons between the fractional dose contributions (in percent) from photon and neutron are taken into account in the determination of the total effective dose. The comparison with mathematical the MIRD phantom and the VIPMAN voxel phantom show partial agreement for neutron effective dose calculations and huge differences for organs absorbed doses. The differences between the three phantoms neutron exposure simulations are discussed and further limitations of voxel-based tomographic phantom are investigated. The future work at the end of this thesis presents a voxel-based eye phantom for high energy proton therapy with initial simulation of a proton beam.

This thesis describes a new design method for a radio frequency quadrupole (RFQ), and its application to the first stage of acceleration for carbon ions in the PAMELA injector. Radiotherapy is a valuable form of cancer treatment, but current methods using photons or electrons make it difficult to deliver an adequate dose to the tumour without damaging healthy surrounding tissue and organs. Charged hadron beams, such as protons and carbon, deposit most of the dose at the Bragg peak, which can be aligned with the tumour. This allows higher doses to treat the cancer while minimising damage to healthy surrounding tissue and organs. The PAMELA project (part of the BASROC consortium) aims to design new charged particle therapy (CPT) facilities using non-scaling fixed-field alternating-gradient accelerators (ns-FFAGs). This new technology offers significant advantages over both cyclotrons and synchrotrons for CPT. The injector for the PAMELA FFAG accelerator includes separate pre-acceleration chains for protons and carbon ions, culminating in a shared injection system into the first FFAG ring. Carbon ions are pre-accelerated by an RFQ and a short linear accelerator (linac). This thesis details the creation of an integrated system of software packages and custom code, which facilitates the design of RFQ vane tips, utilising computer-aided design (CAD) models for both simulation and manufacture, accurate multi-physics modelling of the electric field and particle tracking simulations. This design process is described, along with benchmark results for the Front- End Test Stand (FETS ) RFQ and application of the code in optimising a new RFQ design for PAMELA.