Use this URL to cite or link to this record in EThOS:
Title: Stochastic modelling of carcinogenesis : theory and application
Author: Li, Guangquan
ISNI:       0000 0001 3609 3024
Awarding Body: Imperial College London (University of London)
Current Institution: Imperial College London
Date of Award: 2007
Availability of Full Text:
Full text unavailable from EThOS. Restricted access.
Please contact the current institution’s library for further details.
Cancer is a group of diseases characterized by autonomous, uncontrollable cell proliferation, evasion of cell death, self-construction of oxygen and nutrient supply and spreading of cancerous cells through metastasis. It is vital to elucidate pathways that underlie the cancer process. Mechanistic cancer models, to this extent, attempt to relate the occurrence: of malignant neoplasm to diverse risk factors such as genetic alterations, susceptibility of individuals and exogenous and endogenous carcinogenic exposures. The main objectives of this thesis are to examine the validity of the two-hit hypothesis for retinoblastoma (Knudson, 1971, PNAS (68) 820-23), to unify multiple types of genomic instability and to further assess the role of genomic instability played in the process of carcinogenesis. I shall also explore characteristics of existing mechanistic cancer models. This thesis begins with a survey' of basic cancer biology and existing mechanistic models. Utilizing a fully-stochastic two-stage clonal expansion model, the thesis specifically assesses the validity of the two-hit theory for retinoblastoma-(RB), a childhood ocular malignancy. Comparison of fits of a variety of models (in particular those with up to three mutations) to a population-based RB dfltaset demonstrates the superior fit of the two-stage model to others. This result strongly suggests both the necessity and sufficiency of the two RBI mutations to initiate RB and hence validates the two-hit theory. The thesis goes on to develop a comprehensive,. framework to incorporate multiple types of genomic instability, characterized by-numerous numerical and structural damages exhibiting in the cancer cell genome. This generalized model embraces���·;most, if not all, of the existing MVK-type models. Specific forms of the model are fitted to U.S. white American colon cancer incidence data. Based on comparison of fits to the population-based data, there is little evidence to support the hypothesis that models with more than one type of genomic instability fits better than those with a single type of genomic instability. Since the age-specific incidence data may not possess sufficient information for model discrimination, further investigation is required. The remainder ofthis thesis is concerned with two theoretical aspects. To facilitate a Bayesian implementation for data fitting, a flexible blocking algorithm is developed. In the presence of parameter correlation, the algorithm considerably improves the performance of the Markov chain Monte Carlo simulations. In addition, following a similar approach of Heidenreich et al. (1997, Risk Anal. (17) 391-399), the maximum number of identifiable parameters in the proposed cancer model WIth r types of genomIc InstabIlIty IS r +1 less than the number of biologically-based parameters.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available