Abstract Abstract #6085 Introduction
 A locus on chromosome 8q24.21 has been shown to harbours variants that predispose to breast cancer, colorectal cancer and prostate cancer. The finding of risk variants clustering at 8q24 suggests that there may be common molecular mechanisms that predispose to the common epithelial cancers. The aim of this study was to determine whether the colorectal and prostate susceptibility alleles on 8q24 and other recently identified colorectal and prostate susceptibility alleles elsewhere in the genome are also associated with breast cancer risk.
 Methods
 In this study, we have studied three SNPs on 8q24.21: the breast cancer susceptibility SNP, rs13281615; the colorectal/prostate cancer susceptibility SNP, rs6983267; and the prostate susceptibility SNP, rs16901979; and eleven other SNPs outside of this region that have recently been shown to predispose to either prostate or colorectal cancer. Each SNP was genotyped in 988 sporadic breast cancer cases and 1016 controls from the West of Ireland using KASPar genotyping technology. We combined our data with publicly available datasets (breast: CGEMS, Fletcher et al.; prostate: CGEMS, Eeles et al.; CRC: Tomlinson et al.) by allelic, genotypic, dominant and recessive models using standard techniques of meta-analysis for SNPs previously identified to confer an increased cancer risk in specific tissue types. All analyses were done in Stata 9.2 (Stata Corp, TX).
 Results
 rs13281615 was associated with breast cancer (Pallelic test= 1.8 -2) with an odds ratio of 1.17. However there was no association between any of the prostate or colorectal susceptibility SNPs at 8q24 or elsewhere with breast cancer risk in our cohort. The data support the existence of a low-penetrance susceptibility locus for breast cancer near rs13281615; this locus seems to be distinct from the prostate and colorectal cancer loci in that region, even if the functional effects of the predisposition loci affect the same target gene or protein. Meta-analysis did not show an association between reported cancer SNPs and cancers in other tissue types.
 Conclusions
 This study suggests that low penetrance susceptibility SNPs for breast, colorectal and prostate cancer are distinct and argues against the presence of ''general' cancer predisposition SNPs. Although 8q24 harbours variants that predispose to all three cancers the susceptibility loci within the region appear to be specific for the different cancer types. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6085.