Genome-wide association study identifies novel breast cancer susceptibility loci

Abstract

Breast cancer exhibits familial aggregation, consistent with variation in genetic susceptibility to the disease. Known susceptibility genes account for less than 25% of the familial risk of breast cancer, and the residual genetic variance is likely to be due to variants conferring more moderate risks. To identify further susceptibility alleles, we conducted a two-stage genome-wide association study in 4,398 breast cancer cases and 4,316 controls, followed by a third stage in which 30 single nucleotide polymorphisms (SNPs) were tested for confirmation in 21,860 cases and 22,578 controls from 22 studies. We used 227,876 SNPs that were estimated to correlate with 77% of known common SNPs in Europeans at r2 > 0.5. SNPs in five novel independent loci exhibited strong and consistent evidence of association with breast cancer (P < 10(-7)). Four of these contain plausible causative genes (FGFR2, TNRC9, MAP3K1 and LSP1). At the second stage, 1,792 SNPs were significant at the P < 0.05 level compared with an estimated 1,343 that would be expected by chance, indicating that many additional common susceptibility alleles may be identifiable by this approach.

Figure 1. Quantile–quantile plots for the test statistics (Cochran-Armitage 1 d.f. χ² trend tests) for stages 1 and 2

a, Stage 1; b, stage 2. Black dots are the uncorrected test statistics. Red dots are the statistics corrected by the genomic control method (λ=1.03 for stage 1, λ=1.06 for stage 2). Under the null hypothesis of no association at any locus, the points would be expected to follow the black line.

Figure 2. Forest plots of the per-allele odds ratios for each of the five SNPs reaching genome-wide significance

a, rs2981582; b, rs3803662; c, rs889312; d, rs13281615; and e, rs3817198. The x-axis gives the per-allele odds ratio. Each row represents one study (see Supplementary Table 2), with summary odds ratios for all European and all Asian studies, and all studies combined. The area of the square for each study is proportional to the inverse of the variance of the estimate. Horizontal lines represent 95% confidence intervals. Diamonds represent the summary odds ratios, with 95% confidence intervals, based on the stage 3 studies only.

Figure 3. The FGFR2 locus

a, Map of the whole FGFR2 gene, viewed relative to common SNPs on HapMap. The gene is 126 kb long and in reverse 3′–5′ orientation on chromosome 10. Exon positions are illustrated with respect to the 67 SNPs with m.a.f.>5% in HapMap CEU (therefore the map is not to physical scale). Numbered SNPs are those tested in the genomewide study. SNPs in black were not significant in stage 1. Those in red were significant at P<0.0001 after stage 2. rs10510097 (orange) was significant in stage 1, but failed quality control in stage 2 owing to deviation from Hardy–Weinberg equilibrium. Squares indicate pairwise r² on a greyscale (black=1, white=0). Red circle indicates rs2981582. b, Resequenced 32 kb region, shown relative to SNPs in CEU with m.a.f.>5%, showing pairwise LD for SNPs in HapMap CEU (left panel) and JPT/CHB (right panel). Red circle indicates rs2981582, shown in bold black. c, Sequence conservation of 32 kb region in five species, relative to human sequence (http://pipeline.lbl.gov/methods.shtml). Red circle indicates rs2981582. SNPs in grey are those used in the initial tagging of known common HapMap SNPs within the block. SNPs in black are correlated with rs2981582 with r²>0.6 in European samples. Six SNPs in red were those consistent with being the causative variant on the basis of the genetic data (not excluded at odds of 100:1 relative to the SNP with the strongest association, rs7895676).