Association of ESR1 Germline Variants with TP53 Somatic Variants in Breast Tumors in a Genome-wide Study

Abstract

In breast tumors, somatic mutation frequencies in TP53 and PIK3CA vary by tumor subtype and ancestry. Emerging data suggest tumor mutation status is associated with germline variants and genetic ancestry. We aimed to identify germline variants that are associated with somatic TP53 or PIK3CA mutation status in breast tumors. A genome-wide association study was conducted in 2,850 women of European ancestry with breast cancer using TP53 and PIK3CA mutation status (positive or negative) as well as specific functional categories [e.g., TP53 gain-of-function (GOF) and loss-of-function, PIK3CA activating] as phenotypes. Germline variants showing evidence of association were selected for validation analyses and tested in multiple independent datasets. Discovery association analyses found five variants associated with TP53 mutation status with P values <1 × 10-6 and 33 variants with P values <1 × 10-5. Forty-four variants were associated with PIK3CA mutation status with P values <1 × 10-5. In validation analyses, only variants at the ESR1 locus were associated with TP53 mutation status after multiple comparisons corrections. Combined analyses in European and Malaysian populations found ESR1 locus variants rs9383938 and rs9479090 associated with the presence of TP53 mutations overall (P values 2 × 10-11 and 4.6 × 10-10, respectively). rs9383938 also showed association with TP53 GOF mutations (P value 6.1 × 10-7). rs9479090 showed suggestive evidence (P value 0.02) for association with TP53 mutation status in African ancestry populations. No other variants were significantly associated with TP53 or PIK3CA mutation status. Larger studies are needed to confirm these findings and determine if additional variants contribute to ancestry-specific differences in mutation frequency.

Significance: Emerging data show ancestry-specific differences in TP53 and PIK3CA mutation frequency in breast tumors suggesting that germline variants may influence somatic mutational processes. This study identified variants near ESR1 associated with TP53 mutation status and identified additional loci with suggestive association which may provide biological insight into observed differences.

FIGURE 1

Variant selection for validation studies. Flow chart for variant prioritization for validation studies is shown for TP53 mutation associations (A) and PIK3CA mutation associations (B). Variants were first filtered by P values, MAF, and location near relevant GWAS loci. Variants were then filtered by ORs, location relative to gene with role in breast cancer, TP53 or PI3K/AKT pathway and then by RegulomeDB score. Variants are only counted once but may fall within one or more categories.

FIGURE 2

Manhattan plots for TP53 and PIK3CA discovery GxM analyses. Discovery GWAS data for 879 TP53 mutation carriers and 1,965 breast cancer cases without TP53 mutations (A), 237 cases with TP53 GOF mutations and 1,965 breast cancer cases without TP53 mutations (B), 536 cases with TP53 LOF mutations and 1,965 controls (C), 1,095 PIK3CA mutation carriers and 1,642 breast cancer cases without PIK3CA mutations (D), and 858 cases with PIK3CA activating/hotspot mutations and 1,642 breast cancer cases without PIK3CA mutations, are plotted by −log₁₀ (P values; E). Blue lines represent P values of less than 1 × 10⁻⁵. Chromosome numbers are indicated. GXM, germline variant by mutation; GWAS, genome-wide association study; GOF, gain of function; LOF, loss of function.