Breast cancer (BrCa) is the most frequently diagnosed cancer in women in the US (prevalence 2.6 million), poses a lifetime risk of 1 in 8, and has a death rate of ~40,000/year. About 10% of BrCa clusters in families (260,000 US women);however, >75% of heritable factors are unknown. The 'missing heritability'is thought to be, in part, due to rare inherited susceptibility variants. Our observations using a unique statewide genealogical and cancer database are consistent with this. Large high-risk pedigrees are accepted to be enriched for rare risk variants and a relevant study design to pursue them. The discovery of rare risk variants is important because they may have immediate relevance for the women carrying them. Furthermore, we hypothesize that more common, low- and moderate-risk variants will also exist in the genes identified. Hence, the identification of novel rare risk variants has the potential to provide immediate clinical impact as well as insight into susceptibility at the population-level and new directions for the field. Two major obstacles hinder rare risk variant discovery: a lack of powerful statistical methods for large pedigrees to identify genomic regions of importance;and genetic heterogeneity -both of which are especially challenging for common diseases, such as BrCa. Recently, we introduced shared genomic segment (SGS) methods which have been shown through simulation and proof-of-principle examples to have good power to identify chromosomal regions harboring rare risk variants in high-risk pedigrees. A regionally-guided strategy allows for efficient, focused sequencing and analysis efforts and strong filtering based on sharing. If, however, compelling regions cannot be identified, whole exome sequencing is a complementary strategy. Whether regional or whole exome, it is important that the cases sequenced in a pedigree share the underlying susceptibility variant/s. Membership in a high-risk pedigree certainly increases the likelihood for homogeneity;however, it is expected that other sources of heterogeneity will exist. We hypothesize that molecular tumor subtyping will be critical to identify more genetically homogeneous BrCa. Breast tumors are known to be heterogeneous in cellular and molecular make-up, and that this molecular heterogeneity can be reduced to four major tumor subtypes that have similar gene expression profiles based on 'intrinsic'gene sets. We will study 25 high-risk intrinsic-characterized BrCa pedigrees to identify rare BrCa risk variants and three follow-up cohorts (3,500 individuals) to identify lower-risk variants. Our massively-parallel sequencing efforts will follow a two-pronged approach, including regionally- guided and whole exome sequencing based on the existence of compelling shared regions. Our follow-up in independent cohorts is imperative for corroboration and to establish a broader understanding of how novel BrCa susceptibility genes more generally influence risk. In summary, we believe our research strategy is well- grounded, highly innovative and a powerful approach to identify new BrCa susceptibility variants.
The ability to study large high-risk breast cancer pedigrees, in addition to our incorporation of novel analysis techniques and tumor subtyping, uniquely positions us to identify rare inherited BrCa susceptibility variants. The discovery of such variant could lead to the identification of additional common, low- and moderate-risk genetic factors and improved screening procedures in high-risk families and in the general population.
|Lin, Wei-Yu; Brock, Ian W; Connley, Dan et al. (2013) Associations of ATR and CHEK1 single nucleotide polymorphisms with breast cancer. PLoS One 8:e68578|