Discovery of Novel Breast Cancer Susceptibility Genes by Whole-Exome Sequencing of Affected Sisters

Gesseca Gos1,2, Shelley Bull1, Irene Andrulis1,2

1. Lunenfeld-Tanenbaum Research Institute, Toronto, Canada; 2. Department of Molecular Genetics, University of Toronto, Toronto, Canada

The identification of genes contributing to breast cancer susceptibility is essential to our understanding of the molecular basis of breast cancer. A greater understanding of the biology behind this disease can improve screening, diagnosis, and treatment for patients and their families. Causal mutations previously identified in susceptibility genes account for only one third of the observed heritability, and therefore additional genes contributing to breast cancer susceptibility remain undetected. Recent advances in high-throughput DNA sequencing technology has created new opportunities to interrogate large numbers of genes that may play a role in cancer risk, especially those of rare population frequencies and moderate penetrance that cannot be identified by traditional linkage methods or DNA association studies.

We performed whole-exome sequencing at 50X coverage on blood samples of 42 pairs of sisters affected by breast cancer, at least one of which had an early-onset diagnosis (before 46 years of age). To increase the likelihood of discovering novel genetic variants, the sister pairs were selected from families confirmed to be negative for BRCA1/2 and CHEK2*1100delC mutations. The families are members of the Ontario site of the Breast Cancer Family Registry, an international consortium of 3,500 families that represent a continuum of risk for breast cancer.

Through analysis of the exome sequences, using the Broad Institute’s Genome Analysis Toolkit, and accompanying bioinformatics tools, we identified a short list of 15 genes containing predicted loss-of-function mutations that are rare in the general population, according to the 1000 Genomes Project and NHLBI Exome Sequencing Project (minor allele frequency <0.05), and occur in both sisters of at least two of the sister-pairs in our sample.

Currently we are investigating the rare nonsynonymous mutations in these genes among the cohort, with the aim of choosing several of particular interest, based on their frequencies in the sample and their expected effects on protein functions according to variant effect prediction tools (SIFT, PolyPhen-2, MutationTaster, and MutationAssessor), as candidates for a case-control mutation screening study across a larger sample of affected and non-affected individuals in the Breast Cancer Family Registry. Our objective is to establish a stronger relationship between the mutations in the genes of interest and breast cancer susceptibility.