Ovarian cancer causes more deaths each year among North American women than any other gynecologic cancer. The etiology is poorly understood. Although highly penetrant mutations in BRCA1 and BRCA2 are known to significantly increase ovarian cancer risk, such mutations are rare in the population and collectively they account for only 12-15% of cases. Recent evidence suggests that subtle (non-truncating) but more common genetic variants (i.e. single nucleotide polymorphisms with greater than 5% population frequency) confer more moderate increased risks of cancer and are likely to be involved in a significant proportion of cases. However, most studies conducted to date have tended to rely extensively on candidate gene approaches. Because current understanding of the pathobiology of ovarian cancer is limited, selection of appropriate candidates is challenging and efforts to date have proven largely unsuccessful. Our hypothesis is that ovarian cancer susceptibility genes exist but that the most fruitful strategy for their identification is a genome-wide analysis. Recent methodological and technical developments make this possible and feasible. Our approach will be to combine the resources and expertise of four large case control studies of ovarian cancer that have collected genomic DMA and relevant risk factors on participants. Phase I entails the use of 366,722 haplotype-tagging (ht) SNPs to screen the entire genome for potential loci of interest using 367 cases with a family history of ovarian cancer and 479 matched controls. Phase II seeks to validate and refine these results among 303 population-based cases and 303 matched controls using the 13,000 SNPs most strongly associated with risk. In Phase III we will assess the reproducibility and generalizability of our results among 3072 cases and 3072 matched controls, using the top 7309 SNPs from Phase II. Success in this endeavor will not only allow identification of women at risk for ovarian cancer, but will elucidate genes involved in the pathogenesis of this deadly malignancy.
Showing the most recent 10 out of 49 publications
