2. Single Nucleotide Polymorphism (SNP)-Array Platforms to Assess Breast Cancer Risk
A significant proportion of breast cancers arise in a subset of women who have multiple affected relatives as a result of inherited genetic factors that increase the risk of developing the disease. The relative risk (RR) of breast cancer in mothers and sisters of patients is increased, ranging from 1.8-fold to more than 5-fold [10,11]. In 5%–10% patients, inherited mutations in highly penetrant cancer susceptibility genes, such as BRCA1 and BRCA2, are known to confer a significantly elevated risk (>10-fold) of breast cancer and their carrier relatives [12]. A further 5% of cases carry deleterious variants in moderate-risk breast cancer susceptibility genes, such as CHEK2, ATM, BRIP1, and PALB2 [11,12,13,14]. However, these variants are too rare to be identified in most genome-wide association studies and do not increase risk sufficiently for capture by linkage analysis in family studies.
Numerous genome-wide association studies for different population groups have successfully been performed to discover low-risk SNP variants that are associated with breast cancer [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33]. Such studies have been underpinned by SNP array platforms from companies, such as Affymetrix, Illumina and Perlegen Sciences, ranging in genome coverage, spatial resolution and design. Probes used on SNP arrays for these studies have generally been selected to target SNPs with a minor allele frequency greater than 5%. Thus, genome-wide association studies are designed to detect causal variants that are relatively common in the population. As breast cancer studies have grown in size, less common variants are able to be assessed for risk association. A recent initiative as part of the Collaborative Oncological Gene-Environment Study (COGS) used a custom-designed array to assess almost 200,000 SNPs across the genome in approximately 50,000 breast cancer cases and 50,000 controls [28]. Studies of this size are statistically powered to evaluate variants with a minor allele frequency <5%. As a result of the large COGS initiative, more than 90 independent common susceptibility loci have now been identified, explaining a further 16% of the familial risk [27].
Currently known low-, moderate- and high-risk genetic factors explain up to half of the familial clustering in breast cancer [28]; thus, for a substantial fraction of women, the genetic changes contributing to breast cancer remains undetermined, even if they have a family history [34]. Discovery of variants to explain this “missing heritability” is of clinical relevance, but will require different approaches that perhaps include other types of genetic variation, such as CNVs, using high throughput technology.