Breast tumor DNA methylation patterns associated with smoking in the Carolina Breast Cancer Study

Abstract

Purpose: Tobacco smoking is a risk factor in several cancers, yet its roles as a putative etiologic exposure or poor prognostic factor in breast cancer are less clear. Altered DNA methylation contributes to breast cancer development and may provide a mechanistic link between smoking and gene expression changes leading to cancer development or progression.

Methods: Using a cancer-focused array, we examined methylation at 933 CpGs in 517 invasive breast tumors in the Carolina Breast Cancer Study to determine whether methylation patterns differ by exposure to tobacco smoke. Multivariable generalized linear regression models were used to compare tumor methylation profiles between smokers and never smokers, overall, or stratified on hormone receptor (HR) status.

Results: Modest differences in CpG methylation were detected at p < 0.05 in breast tumors from current or ever smokers compared with never smokers. In stratified analyses, HR- tumors from smokers exhibited primarily hypomethylation compared with tumors from never smokers; hypomethylation was similarly detected within the more homogeneous basal-like subtype. Most current smoking-associated CpG loci exhibited methylation levels in former smokers that were intermediate between those in current and never smokers and exhibited progressive changes in methylation with increasing duration of smoking. Among former smokers, restoration of methylation toward baseline (never smoking) levels was observed with increasing time since quitting. Moreover, smoking-related hypermethylation was stronger in HR+ breast tumors from blacks than in whites.

Conclusions: Our results suggest that breast tumor methylation patterns differ with tobacco smoke exposure; however, additional studies are needed to confirm these findings.

Keywords: Breast cancer; Breast tumor; Epigenetic; Methylation; Smoking.

Fig. 1

Patterns of breast tumor differential methylation in smokers compared with never smokers. Generalized linear regression models (GLM) (logit link) adjusted for age, race, menopausal status, stage, BMI, and alcohol consumption were used to identify CpG loci differentially methylated in breast tumors in smokers versus never smokers. Volcano plots display array-wide patterns of breast tumor differential methylation in current or ever smokers among: a all cases, b HR+ cases, c HR− cases, d HR− cases according to smoking duration (long-term >20 years or shorter-term ≤20 years) among ever smokers, or e basal-like cases. Each volcano plot displays the negative log of unadjusted p-values for differences in β (proportion DNA methylated) at each probe on the y axis versus the correlation coefficient for methylation at each CpG locus on the x axis. Probes that fall above the broken line are significant at p <0.05. Probes hypomethylated in smokers have negative coefficients, while probes hypermethylated in smokers have positive coefficients. f Bar graph summarizing numbers of differentially hypomethylated or hypermethylated CpG probes at p < 0.05 in ever smokers, current smokers, long-term (>20 year), or shorter-term (<20 year) smokers versus never smokers among all cases or by HR status

Fig. 2

Boxplots illustrating methylation levels at top CpGs differing in methylation between current and never smokers with HR− breast tumors. a Methylation differences according to smoking status in current (C) (n = 37) or former (F) (n = 34) smokers compared with never smokers (N) (n = 92). b Methylation with increasing duration of smoking in years among ever smokers: N (n = 92), ≤10 (n = 25), 11–20 (n = 16), or >20 years (n = 29). c Methylation with years since quitting among former smokers: C (n = 37), ≤10 (n = 18), 11+ (n = 16), N (n = 92). Never or current smokers are included for reference in some plots as appropriate. Boxplot boundaries indicate the interquartile range, and mean and median are indicated by the black bar and diamond, respectively

Fig. 3

Patterns of breast tumor differential methylation in black or white current smokers versus never smokers with HR+ or HR− breast cancer. Generalized linear regression (GLM) models adjusted for age, race, menopausal status, stage, BMI, and alcohol consumption were used to compare breast tumor methylation beta values between current and never smokers. a Volcano plots display array-wide patterns of breast tumor differential methylation in current versus never smokers among a black HR− cases, b black HR+ cases, c white HR− cases, and d white HR+ cases. Probes that fall above the broken line are significant at p <0.05. Probes hypomethylated in smokers have negative coefficients and probes hypermethylated in smokers have positive coefficients. e Bar graph summarizing numbers of differentially hypomethylated or hypermethylated CpG probes in black and white current smokers with HR− or HR+ breast tumors. f Venn diagrams summarizing the overlap of CpG probes differentially methylated in current smokers among black and white cases with HR− or HR+ breast cancers. g Boxplots of selected smoking-related CpG loci illustrating the range of beta values in HR− breast tumors from black or white current or former smokers versus never smokers