Germline DNA variants as determinants for breast cancer predisposition and prognosis

Abstract

Breast cancer is the most common cancer among women in the developed world. The disease results from the combined effects of genetic, environmental, reproductive and lifestyle risk factors. Germline DNA variations identified thus far by linkage and genome-wide association studies (GWASs) account for less than one-third of variability in breast cancer predisposition, suggesting that more variants exist. Furthermore, despite advancements in breast cancer therapies guided by tumor-based prognostic and predictive factors, approximately 30% of breast cancer patients who receive standard guideline-based therapies experience disease recurrence within ten years post diagnosis. Consequently, there is a clear need of additional markers for disease risk assessment as well as markers of potential prognostic values to better guide treatment modalities. In this thesis, I adopted a comprehensive approach utilizing single nucleotide polymorphisms (SNPs) and germline copy number aberrations (copy number variations (CNVs) and copy neutral-loss of heterozygosities (CN-LOHs)) to identify markers for breast cancer susceptibility and disease prognosis. I used a multi-stage association study design that included cumulative sample sizes of 2,795 invasive breast cancer cases and 4,505 healthy controls of predominantly Caucasian in origin selected from Alberta, Canada. I identified a novel breast cancer susceptibility locus on chromosome 4q31.22 showing a strong statistical significance for overall breast cancer (per allele odds ratio=1.28 and P=1.5 x 10-7), adjusted for body mass index (BMI). I also independently confirmed one literature reported association on chromosome 8q24.21-rs13281615 (BMI adjusted-P<3.1 x 10-3) with breast cancer prognosis. Since epistatic interactions have been hypothesized to capture additional heritability for breast cancer, I extended my studies and identified interactions involving two SNPs and an interaction involving four SNPs. These interactions were from the single-locus effects with weak statistical significance in GWAS and/or candidate-gene studies. Finally, I identified germline CNAs as potential prognostic markers for the predominant luminal A breast cancers (up to 70% of total cases diagnosed), which recur despite the good prognosis. Germline DNA-based markers for disease predisposition and prognosis is an area in its infancy and clearly more work is warranted to substantiate and extend the reported findings to enable eventual translation of research to clinical applications.