Admixture mapping identifies a locus on 6q25 associated with breast cancer risk in US Latinas

Abstract

Among US Latinas and Mexican women, those with higher European ancestry have increased risk of breast cancer. We combined an admixture mapping and genome-wide association mapping approach to search for genomic regions that may explain this observation. Latina women with breast cancer (n= 1497) and Latina controls (n= 1272) were genotyped using Affymetrix and Illumina arrays. We inferred locus-specific genetic ancestry and compared the ancestry between cases and controls. We also performed single nucleotide polymorphism (SNP) association analyses in regions of interest. Correction for multiple-hypothesis testing was conducted using permutations (P(corrected)). We identified one region where genetic ancestry was significantly associated with breast cancer risk: 6q25 [odds ratio (OR) per Indigenous American chromosome 0.75, 95% confidence interval (CI): 0.65-0.85, P= 1.1 × 10(-5), P(corrected)= 0.02]. A second region on 11p15 showed a trend towards association (OR per Indigenous American chromosome 0.77, 95% CI: 0.68-0.87, P= 4.3 × 10(-5), P(corrected)= 0.08). In both regions, breast cancer risk decreased with higher Indigenous American ancestry in concordance with observations made on global ancestry. The peak of the 6q25 signal includes the estrogen receptor 1 (ESR1) gene and 5' region, a locus previously implicated in breast cancer. Genome-wide association analysis found that a multi-SNP model explained the admixture signal in both regions. Our results confirm that the association between genetic ancestry and breast cancer risk in US Latinas is partly due to genetic differences between populations of European and Indigenous Americans origin. Fine-mapping within the 6q25 and possibly the 11p15 loci will lead to the discovery of the biologically functional variant/s behind this association.

Figure 1.

Results of breast cancer admixture mapping. On the X-axis are genomic positions by chromosome. On the Y-axis are the negative log₁₀ P-values for the association between locus-specific ancestry and breast cancer risk. (A) Admixture mapping for Indigenous American component. (B) Admixture mapping for European component. (C) Admixture mapping for Africa component.

Figure 2.

Results of ER subtype-specific breast cancer admixture mapping for the Indigenous American component. On the X-axis are genomic positions within each chromosome. On the Y-axis are the negative log₁₀ P-values for the association between locus-specific ancestry and ER-specific breast cancer subtype. Results for ER-positive analyses are in blue and for ER-negative analyses are in red.

Figure 3.

Admixture mapping and SNP association results. The solid line represents the admixture mapping signal. The blue dots represent the negative log₁₀ P-values of association between single SNPs and breast cancer risk (all SNPs that showed an association with breast cancer risk at a P-value ≤ 0.05 were included). The red dots represent the locus-specific ancestry association after adjustment for selected individual SNPs (SNPs that showed an association with breast cancer at a P-value ≤ 0.001). The blue dots with black hollow squares are the SNPs that modified the locus-specific ancestry association signal represented by the red dots. (A) 6q25 region. (B) 11p15 region.