Feasibility of selective multiple boosting in the planning of intracranial radiation therapy

Walker, Charlotte Anne

Clinical biosciences; Radiotherapy
August 2007

Thesis or dissertation


Rights
© 2007 Charlotte Anne Walker. All rights reserved. No part of this publication may be reproduced without the written permission of the copyright holder.
Abstract

IMRT provides unprecedented means to sculpt radiation dose in three-dimensions, pledging potential to improve local disease control via conformal dose escalation as well as better normal tissue sparing via conformal avoidance. In this thesis, the feasibility and practicality of delivering high doses to intelligently-defined multiple tumour sub-volumes is verified for intracranial radiation treatments. The term ‘selective multiple boosting’ (SMB) is adopted for the proposed planning scheme. Physical control over local dose deposition is characterised and quantified through the design of two pseudo-anatomy models. The models show that intra-structural optimisation is easily implemented within a standard IMRT planning module. For concentric, spherical boost volumes, regional dose can be controlled at 10 and 5 mm resolution, where a dose differential of 5 and 3 Gy respectively is readily achievable, whilst incrementally boosting neighbouring volumes is more difficult. The limitations of functional imaging techniques are discussed in this context and resolution issues investigated for magnetic resonance spectroscopic imaging (MRSI). Interpolation experiments show that coarse resolutions are not a barrier to using data effectively for defining boost volumes. However, a need for caution and further research on the interpretation and reproducibility of advanced imaging modalities is highlighted. The idea of a composite or hybrid target volume is presented, consisting of multiple tumour contours delineated on different imaging sequences and amalgamated into a probability map of tumour existence. Software has been written for this purpose and demonstrated viable for SMB planning on real patient data. An IMRT plan evaluation toolbox has also been developed, implementing both existing and novel means of characterising the dosimetry and biological effect of these advanced and complex plans. Whilst it is recognised that further work and understanding is required, it is concluded that SMB is not only feasible but also a valuable pursuit in terms of potential clinical gain.

Publisher
Postgraduate Medical Institute, The University of Hull
Supervisor
Beavis, Andy; Liney, Gary; Phillips, Roger, computer scientist
Sponsor (Organisation)
Institute of Clinical Biosciences, University of Hull; The Princess Royal Hospital (Hull and East Yorkshire Hospitals NHS Trust)
Qualification level
Doctoral
Qualification name
PhD
Language
English
Identifier
hull:578
QR Code