Use this URL to cite or link to this record in EThOS:
Title: Exploring the genetic landscape of breast cancer risk & prognosis
Author: Politopoulos, Ioannis
ISNI:       0000 0004 2741 0341
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2012
Availability of Full Text:
Full text unavailable from EThOS.
Please contact the current institution’s library for further details.
The genetic factors known to be involved in breast cancer risk comprise more than 30 loci. These include the high penetrance early-onset breast cancer genes, BRCA1 and BRCA2, a number of rare cancer syndrome genes and rare genes with more moderate penetrance. A larger group of common variants has more recently been identified through genome-wide association studies. Collectively the breast cancer loci identified to date contribute only -30% of the familial risk. In this study, the most prevalent method of conducting genome-wide analyses - single marker testing - was compared to a composite likelihood approach testing multiple markers at the same time based on the underlying linkage disequilibrium structure. Evidence indicates increased power of detection of variants genome-wide for the latter. In order to increase power of detection, three published genome-wide datasets were combined in a meta-analysis using the composite likelihood approach. Findings demonstrate replication of well-established breast cancer loci [such as FGFR2and 8q24] and identify a novel candidate gene, PIK3AP1, as a highly ranking locus. Additionally, part of this study focused on oestrogen receptor-negative [ER -] breast cancer risk by combining genotypic data from seven case-control cohorts and performing composite likelihood testing. Despite the limited replication of known breast cancer loci, which can be justified by the prior prevalence of oestrogen receptor positive [ER+] studies, a nominally significant locus in lOq24 shows association with the ER- form of the disease. Evidence also suggests alternative adjacent locations for the causal variants of previously reported breast cancer loci, such as COX11 and TOX3 genes. Ultimately, genome-wide marker analysis focused on early-onset breast cancer risk and prognosis revealed the confounding effects of combining data genotyped separately without recalling genotypes based on the raw intensity files. However, the association of rs10749071 with HER2+ [Human Epidermal Growth Factor Receptor 2-positive] early onset breast cancer and rs3764459 with the survival of anthracycline-treated breast tumours replicated in follow-up studies. The novel loci identified need to be further tested in larger independent cohorts before conclusive arguments are drawn.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available