Consortium analysis of 7 candidate SNPs for ovarian cancer

Susan J Ramus, Robert A Vierkant, Sharon E Johnatty, Malcolm C Pike, David J Van Den Berg, Anna H Wu, Celeste Leigh Pearce, Usha Menon, Aleksandra Gentry-Maharaj, Simon A Gayther, Richard A DiCioccio, Valerie McGuire, Alice S Whittemore, Honglin Song, Douglas F Easton, Paul D P Pharoah, Montserrat Garcia-Closas, Stephen Chanock, Jolanta Lissowska, Louise BrintonKathryn L Terry, Daniel W Cramer, Shelley S Tworoger, Susan E Hankinson, Andrew Berchuck, Patricia G Moorman, Joellen M Schildkraut, Julie M Cunningham, Mark Liebow, Susanne Krüger Kjaer, Estrid Hogdall, Claus Hogdall, Jan Blaakaer, Roberta B Ness, Kirsten B Moysich, Robert P Edwards, Michael E Carney, Galina Lurie, Marc T Goodman, Shan Wang-Gohrke, Silke Kropp, Jenny Chang-Claude, Penelope M Webb, Xiaoqing Chen, Jonathan Beesley, Georgia Chenevix-Trench, Ellen L Goode


    The Ovarian Cancer Association Consortium selected 7 candidate single nucleotide polymorphisms (SNPs), for which there is evidence from previous studies of an association with variation in ovarian cancer or breast cancer risks. The SNPs selected for analysis were F31I (rs2273535) in AURKA, N372H (rs144848) in BRCA2, rs2854344 in intron 17 of RB1, rs2811712 5' flanking CDKN2A, rs523349 in the 3' UTR of SRD5A2, D302H (rs1045485) in CASP8 and L10P (rs1982073) in TGFB1. Fourteen studies genotyped 4,624 invasive epithelial ovarian cancer cases and 8,113 controls of white non-Hispanic origin. A marginally significant association was found for RB1 when all studies were included [ordinal odds ratio (OR) 0.88 (95% confidence interval (CI) 0.79-1.00) p = 0.041 and dominant OR 0.87 (95% CI 0.76-0.98) p = 0.025]; when the studies that originally suggested an association were excluded, the result was suggestive although no longer statistically significant (ordinal OR 0.92, 95% CI 0.79-1.06). This SNP has also been shown to have an association with decreased risk in breast cancer. There was a suggestion of an association for AURKA, when one study that caused significant study heterogeneity was excluded [ordinal OR 1.10 (95% CI 1.01-1.20) p = 0.027; dominant OR 1.12 (95% CI 1.01-1.24) p = 0.03]. The other 5 SNPs in BRCA2, CDKN2A, SRD5A2, CASP8 and TGFB1 showed no association with ovarian cancer risk; given the large sample size, these results can also be considered to be informative. These null results for SNPs identified from relatively large initial studies shows the importance of replicating associations by a consortium approach.

    TidsskriftInternational Journal of Cancer
    Udgave nummer2
    Sider (fra-til)380-388
    Antal sider9
    StatusUdgivet - 15 jul. 2008


    Dyk ned i forskningsemnerne om 'Consortium analysis of 7 candidate SNPs for ovarian cancer'. Sammen danner de et unikt fingeraftryk.