Use this URL to cite or link to this record in EThOS:
Title: Weight of evidence analysis in framework approaches to cancer hazard identification : a case study with phenobarbital
Author: Idahosa, Ehi
ISNI:       0000 0004 2699 6395
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis addresses the transparency of weight of evidence-based analysis in the application of the IPCS Framework for Assessing the Human Relevance of a carcinogenic response to a chemical. This thesis also explores how, and to what extent, available toxicogenomics data can be incorporated into a cancer mode of action analysis. A systematic review was carried out on epidemiological data available for phenobarbital (PB). A positive association between liver cancer and PB was reported among a large Finnish cohort however the study was assessed to be unreliable on the basis of key limitations. However owing to the size of cohort investigated, the study could not be disregarded and the conclusions reached by the authors were considered to be associated with a high level of uncertainty. Studies of two smaller cohorts of patients, assessed to be of a high quality, reported no findings of such an association. The methods applied in this thesis illustrated the value of systematic review techniques in increasing the transparency of decisions made when applying a WOE based approach to assessing the human relevance of cancer mode of action and in helping to identify important data gaps. Comparative analysis of publicly available toxicogenomics data for PB using Pathway Analysis software led to the identification of perturbed pathways that were interpreted within the context of the key events for PB MOA for hepatocarcinogenesis. The analysis demonstrated that prior mechanistic knowledge about a compound under investigation is currently necessary before efficient interpretation of the biological relevance of perturbed pathways is possible. To realise the true potential for the application of toxicogenomics data in risk assessment, future toxicogenomics study designs will need to be more targeted (e.g. in their considerations of dose-response, time-response and species differences) before they can be applied to help fill the current gaps in mechanistic knowledge. The experience from the comparative analysis reported here suggests that at the present time, information on perturbed pathways obtained in this way is not sufficiently robust to be used as stand alone evidence in MOA analysis. The potential to read across toxicogenomics evidence for PB MOA to other compounds, which may act via the same MOA however deserves further exploration.
Supervisor: Boobis, Alan ; Rushton, Lesley Sponsor: Department of Health
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral