Next generation sequencing approaches to identify novel susceptibility genes for epithelial ovarian cancer

Ovarian cancer is the fifth most common cancer in women in developed countries and is associated with poor survival due to late diagnoses. Strategies focusing on detecting the disease in the earliest stages and/or improving risk prediction may represent effective clinical intervention reducing disease burden. Women at the greatest risk of epithelial ovarian cancer (EOC) can be offered prophylactic risk-reducing salpingo-oopherectomy (RRSO), which is currently only offered to women with mutations in the highly penetrant susceptibility genes BRCA1 or BRCA2. Previous studies show that 46% of familial cases of EOC carry a deleterious mutation in BRCA1 (37%) or BRCA2 (9%). The residual proportion of familial risk is likely to be attributable to other genetic variants providing a rationale for identifying additional susceptibility alleles using rapid high-throughput next generation sequencing (NGS) in large samples sizes. A pilot study determines the principle of NGS in mutation detection sequencing BRCA1 gene in 12 DNA samples with known mutations. The 11bp deletion, missed in the analysis, is detected by altering the bioinformatics. The second study sequences 1506 cases and 1130 healthy controls using Fluidigm microfluidic technology and Illumina HiSeq2000 in 6 DNA repair genes (RAD51B, RAD51C, RAD51D, XRCC2, XRCC3 and SLX4). 94% of the targeted region is sequenced with >30 reads. 23 cases and 1 control show a putative protein-truncating variant in 5 genes. Many missense variants are detected in cases and controls suggesting these are not pathogenic. Epidemiological data shows that women with family history and a deleterious mutation develop EOC on average 10 years younger. Interestingly, half of those women with detected mutations have no family history. A final study uses the established NGS approach to characterise the mutation prevalence in 4 known and 5 candidate EOC susceptibility genes in 2300 unaffected women from high-risk breast-ovarian cancer families. BRCA1 and BRCA2 deleterious mutations are identified in 53 and 49 women respectively. Deleterious mutations are detected in 6 additional genes, BRIP1 (n=5), RAD51C (n=3), RAD51D (n=1) PALB2 (n=5), BARD1 (n=1) and NBN (n=3). Importantly, a bioinformatics pipeline is refined to maximise variant detection sensitivity with zero false negatives where read depth is >30X. Further large case-control studies are recommended to examine the population frequencies in these novel genes. These studies demonstrate the potential of targeted NGS approaches for population-wide risk prediction and early detection of EOC.