Epigenome-wide association study of breast cancer using prospectively collected sister study samples

Abstract

Background: Previous studies have suggested DNA methylation in blood is a potential epigenetic marker of cancer risk, but this has not been evaluated on a genome-wide scale in prospective studies for breast cancer.

Methods: We measured DNA methylation at 27578 CpGs in blood samples from 298 women who developed breast cancer 0 to 5 years after enrollment in the Sister Study cohort and compared them with a random sample of 612 cohort women who remained cancer free. We also genotyped women for nine common polymorphisms associated with breast cancer.

Results: We identified 250 differentially methylated CpGs (dmCpGs) between case subjects and noncase subjects (false discovery rate [FDR] Q < 0.05). Of these dmCpGs, 75.2% were undermethylated in case subjects relative to noncase subjects. Women diagnosed within 1 year of blood draw had small but consistently greater divergence from noncase subjects than did women diagnosed at more than 1 year. Gene set enrichment analysis identified Kyoto Encyclopedia of Genes and Genomes cancer pathways at the recommended FDR of Q less than 0.25. Receiver operating characteristic analysis estimated a prediction accuracy of 65.8% (95% confidence interval = 61.0% to 70.5%) for methylation, compared with 56.0% for the Gail model and 58.8% for genome-wide association study polymorphisms. The prediction accuracy of just five dmCpGs (64.1%) was almost as good as the larger panel and was similar (63.1%) when replicated in a small sample of 81 women with diverse ethnic backgrounds.

Conclusions: Methylation profiling of blood holds promise for breast cancer detection and risk prediction.

Figure 1.

Manhattan plot of methylation case–cohort association P values. Lower dashed horizontal black line indicates unadjusted P equal to .05; upper red dashed line indicates fale discover rate Q value threshold of 0.05 with 250 CpGs falling above this threshold.

Figure 2.

Centered mean methylation values for 250 differentially methylated CpGs (dmCpGs) and time to diagnosis (TTD). A) Centered mean methylation β values in blood for the 250 dmCpGs for case subjects (blue) and noncase subjects (green) were obtained by subtracting the overall (case subjects + noncase subjects) mean from the mean for case subjects or from the mean for noncase subjects. Thus the vertical distance between green and blue dots represents the difference in means between case subjects and noncase subjects. dmCpGs are ordered from those where case subjects are the most undermethylated relative to noncase subjects on the left to those where case subjects are overmethylated relative to noncase subjects on the right. Black bars at the bottom of the figure indicate dmCpGs at CpG nonisland locations. B) Same centering as panel A but separating case subjects into two groups: case subjects with TTD greater than 1year (blue) and case subjects with TTD within 1 year (red triangle). Case subjects with TTD greater than 1 year tend to have intermediate values between the other two groups more than would be predicted by chance (P = 1.33×10⁻⁹).

Figure 3.

Receiver operating characteristics of Gail score, genome-wide association study (GWAS) single nucleotide polymorphims (SNPs), and methylation. Receiver operating curve (ROC) plot of sensitivity (probability of being classified as a case subject given that one is a case subject) versus 1-Specificity (probability of being classified as a noncase subject given that one is a noncase subject). The straight diagonal line indicates random classification corresponding to an area under the curve (AUC) of 50%. ROC curves to show classification performance for Gail score alone (black) and nine GWAS SNPs alone (blue) were obtained using the full set of women. The red ROC curve shows classification performance for an average of 57 methylation markers evaluated in independent test sets of subjects with 95% confidence intervals represented by gray circles. Combining Gail score and nine GWAS SNPs with methylation data provided only slight improvement to 66.1% (not shown in plot) over using methylation data alone.