Investigating the impact of genomic profiling on breast cancer risk prediction, treatment and outcome
MetadataShow full item record
This item's downloads: 1 (view details)
It has, for many years, been acknowledged that “breast cancer” is an umbrella term for many genetically and immunophenotypically distinct malignancies of the breast. Treatment paradigms are shifting in direct response to increasing understanding of the molecular mechanisms underlying this complex set of diseases. Breast cancer, like all other cancers, is “genetic”; triggered by a pathogenic variant in a tumour suppressor or proto-oncogene, or by pathogenic silencing or up-regulation of such genes by copy number variation, epigenetics or variation of regulatory elements. Such variation may occur at the level of the germline or of the soma. The genetic architecture of breast cancer has been well defined, with susceptibility loci categorised as high-, medium- or low-risk depending on the associated disease penetrance. The aim of this study was to explore the utility of molecular profiling of the germline of patients with breast cancer for known and putative high- and low-risk susceptibility loci, and of the tumours of such individuals, and to investigate the association of germline and somatic variation on risk estimation, phenotype, prognostication, treatment planning and outcome. A mixed methods study was undertaken. The referral patterns of clinicians and uptake of patients with respect to BRCA1/BRCA2 testing, as well as phenotype associated with pathogenic mutations in these high-risk genes were examined by observational cohort studies. Surveillance of these high-risk individuals was assessed by patient surveys. A number of case-control studies were undertaken to identify novel cancer susceptibility loci, or to explore the association of putative loci with disease, by targeted sequencing of the 3’ UTRs and miRNA binding sites of cancer predisposition genes, Sanger sequencing, Sequenome, and Taqman genotyping. Functional assays were undertaken to examine the biological impact of one of the variants of interest identified. A longitudinal cohort study was undertaken to examine the influence of tumour gene expression profiling on therapeutic decision-making in patients with breast cancer. 31 families with confirmed pathogenic variants in BRCA1 or BRCA2 in the West of Ireland were identified. A recurrent large genomic rearrangement in BRCA1 (deletion exons 1-23) was found to occur more frequently in patients in this region than elsewhere in Ireland. Patients with such pathogenic variants may be successfully repatriated to National Breast Screening services for surveillance, as it was confirmed that 87-91% of individuals receiving surveillance through this pathway receive at least recommended screening. A variant in the non-coding region of BRCA1 was also shown to associate with severe disease. A novel variant in the PCM1 gene was identified in patients with ovarian cancer, but did not associate with breast cancer in our cohort. A variant at 12p11 was confirmed to be associated with breast cancer risk in Irish patients. A variant in KRAS was postulated to be associated with disease risk, and was shown to confer differential propensity to colony formation in vitro in response to alterations in the hormonal milieu. It was observed that gene expression analysis of tumours was taken up with high acceptability by Irish patients and clinicians, and treatment decisions were modified accordingly. This work has demonstrated the potential clinical utility of tumour and germline assessment in defining patient risk, prognosis and treatment. However, further larger-scale studies are required to confirm the postulated associations of rare low-penetrance alleles with disease. The potential translational utility of the findings of this work is promising, but increased funding in genetics in the clinical setting is required to ensure patients can avail of genomic profiling where appropriate.