Please use this identifier to cite or link to this item: http://hdl.handle.net/2440/118076
Type: Thesis
Title: Spatio-temporal, multicellular and Monte Carlo track-based model of radiotherapy in silico
Author: Forster, Jake Cameron
Issue Date: 2018
School/Discipline: School of Physical Sciences
Abstract: A notable short-coming in the way radiotherapy is currently practised is that patient-specific radiobiology is minimally and rarely accounted for in the treatment planning process. If this is to be remedied, in silico radiobiological models of radiotherapy will play an essential role. By increasing the complexity of such models, greater accuracy and utility are gained, along with opportunities for new radiobiological insights. A new computational model was developed called “Stochastic Squared Radiotherapy” (S2RT). It is a spatio-temporal/four-dimensional radiotherapy model for head and neck squamous cell carcinoma (HNSCC), that uses stochastic modelling of tumour cells and Monte Carlo track structure simulations. The four main components of the model are tumour growth, tumour irradiation, DNA damage induction and cell death/survival. The tumour growth module generates the initial multicellular tumour and evolves it spatio-temporally in-between dose fractions. Ellipsoidal tumour cells occupy randomised, non-overlapping locations. Cells are pushed outward and fall inward following cell division and cell death, respectively. An epithelial cell hierarchy of stem, transit and differentiated cells is modelled. A connected and chaotic network of blood vessels grows interwoven between the cells. Chronic hypoxia and necrotic cells are simulated at distances far from blood vessels. Hypoxic cells divide slower and necrotic cells are gradually resorbed. Accelerated repopulation may be simulated by increasing the symmetric division of cancer stem cells. Dose fractions are delivered to the tumour in Monte Carlo simulations of radiation tracks. The multicellular tumour is voxelised into nucleus, cytoplasm and intercellular voxels of size 2 μm and imported into Geant4 to perform irradiation. A Geant4 application was developed that uses Geant4-DNA to simulate low-energy physical interactions and radiolytic chemical tracks to account for the indirect effect. The tracks through cell nuclei are converted to DNA damage, including doublestrand breaks (DSBs). This was done by spatially clustering physical interactions such as ionisations and excitations and hydroxyl radical interactions into simulated DNA volumes, each of size 10 base pairs. The DNA damage was made dependent upon the cellular pO2 by increasing the efficiency of DNA radical-to-strand break conversion with increasing pO2. In the model, complex DSBs (cDSBs) produced DNA free-ends that can misrejoin with one another and produce exchange-type chromosome aberrations. Complete exchanges are assumed. The misrejoining probability is modelled as an exponential function of the initial distance between the two cDSBs involved. Cells die if they contain an asymmetric chromosome aberration. Notable findings from the S2RT model include: 1. Symmetric division of cancer stem cells may be as high as 50% during accelerated repopulation. 2. The decrease in the oxygen enhancement ratio for DSB induction with increasing LET can be attributed to spatial clustering alone; i.e., at higher LET, the additional strand breaks produced in the presence of oxygen seldom result in additional DSBs. Instead, they increase DSB complexity. 3. For MV x-rays, misrejoinings between cDSBs produced by the same primary x-ray (including its secondary electrons) do not contribute appreciably to the linear components of chromosome aberration production and cell killing. For HNSCC, which does have an appreciable linear component of cell killing, unrejoined DNA ends (i.e. incomplete exchanges and terminal deletions) may be important. There is promise of accuracy and utility in S2RT because it is predicated on simulating Monte Carlo tracks through a multicellular tumour and simulating cellular tumour growth in-between dose fractions. DNA damage induction and subsequent processes like DNA free-end misrejoining and cell death are modelled stochastically using the track structure. Simulated tumours have realistic spatial distributions of cellular pO2 in relation to the blood vessels, so one can carefully investigate the effect of microscopic regions of tumour hypoxia on treatment efficacy. Since tumour irradiation is performed with track structure, the radiation quality modelled can easily be extended to high LET beams. Modelling a connected network of blood vessels in the tumour also enables consideration of vascular damages. In particular, the model may be used in the future to investigate the extent to which wide-spread vascular damages are responsible for the efficacy of high dose per fraction treatments such as stereotactic body radiotherapy.
Advisor: Bezak, Eva
Phillips, Wendy
Douglass, Michael
Dissertation Note: Thesis (Ph.D.) -- University of Adelaide, School of Physical Sciences, 2018
Keywords: Radiobiology
computational modelling
radiotherapy
Monte Carlo track structure
head and neck cancer
stochastic
Provenance: This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legals
Appears in Collections:Research Theses

Files in This Item:
File Description SizeFormat 
Forster2018_PhD.pdf46.47 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.