This commentary is about predicting a woman’s breast cancer risk from her mammogram, building on the work of Wolfe, Boyd and Yaffe on mammographic density. We summarise our efforts at finding new mammogram-based risk predictors, and how they combine with the conventional mammographic density, in predicting risk for interval cancers and screen-detected breast cancers across different ages at diagnosis and for both Caucasian and Asian women. Using the OPERA (odds ratio per adjusted standard deviation) concept, in which the risk gradient is measured on an appropriate scale that takes into account other factors adjusted for by design or analysis, we show that our new mammogram-based measures are the strongest of all currently known breast cancer risk factors in terms of risk discrimination on a population-basis. We summarise our findings graphically using a path diagram in which conventional mammographic density predicts interval cancer due to its role in masking, while the new mammogram-based risk measures could have a causal effect on both interval and screen-detected breast cancer. We discuss attempts by others to pursue this line of investigation, the measurement challenge that allows different measures to be compared in an open and transparent manner on the same datasets, as well as the biological and public health consequences.

Genome-wide association studies have identified breast cancer risk variants in over 150 genomic regions, but the mechanisms underlying risk remain largely unknown. These regions were explored by combining association analysis with in silico genomic feature annotations. We defined 205 independent risk-associated signals with the set of credible causal variants in each one. In parallel, we used a Bayesian approach (PAINTOR) that combines genetic association, linkage disequilibrium and enriched genomic features to determine variants with high posterior probabilities of being causal. Potentially causal variants were significantly over-represented in active gene regulatory regions and transcription factor binding sites. We applied our INQUSIT pipeline for prioritizing genes as targets of those potentially causal variants, using gene expression (expression quantitative trait loci), chromatin interaction and functional annotations. Known cancer drivers, transcription factors and genes in the developmental, apoptosis, immune system and DNA integrity checkpoint gene ontology pathways were over-represented among the highest-confidence target genes. Fine-mapping of causal variants and integration of epigenetic and chromatin conformation data identify likely target genes for 150 breast cancer risk regions.

We conducted a genome-wide association study of blood DNA methylation and smoking, attempted replication of previously discovered associations, and assessed the reversibility of smoking-associated methylation changes. DNA methylation was measured in baseline peripheral blood samples for 5,044 participants in the Melbourne Collaborative Cohort Study. For 1,032 participants, these measures were repeated using blood samples collected at follow-up, a median of 11 years later. A cross-sectional analysis of the association between smoking and DNA methylation and a longitudinal analysis of changes in smoking status and changes in DNA methylation were conducted. We used our cross-sectional analysis to replicate previously reported associations for current (N=3,327) and former (N=172) smoking. A comprehensive smoking index accounting for the bioactivity of smoking and several aspects of smoking history was constructed to assess the reversibility of smoking-induced methylation changes. We identified 4,496 cross-sectional associations at P<10−7, including 3,296 that were novel. We replicated the majority (90%) of previously reported associations for current and former smokers. In our data, we observed for former smokers a substantial degree of return to the methylation levels of never smokers, compared with current smokers (median: 74%, IQR=63-86%). Consistent with this, we found wide-ranging estimates for the half-life parameter of the comprehensive smoking index. Longitudinal analyses identified 368 sites at which methylation changed upon smoking cessation. Our study provides evidence of many novel associations between smoking and DNA methylation at CpGs across the genome, replicates the vast majority of previously reported associations, and indicates wide-ranging reversibility rates for smoking-induced methylation changes.

Genome-wide association studies (GWAS) have identified more than 170 breast cancer susceptibility loci. Here we hypothesize that some risk-associated variants might act in non-breast tissues, specifically adipose tissue and immune cells from blood and spleen. Using expression quantitative trait loci (eQTL) reported in these tissues, we identify 26 previously unreported, likely target genes of overall breast cancer risk variants, and 17 for estrogen receptor (ER)-negative breast cancer, several with a known immune function. We determine the directional effect of gene expression on disease risk measured based on single and multiple eQTL. In addition, using a gene-based test of association that considers eQTL from multiple tissues, we identify seven (and four) regions with variants associated with overall (and ER-negative) breast cancer risk, which were not reported in previous GWAS. Further investigation of the function of the implicated genes in breast and immune cells may provide insights into the etiology of breast cancer. Over 170 susceptibility loci have been identified by genome-wide association studies in breast cancer. Here, the authors interrogated the role of risk-associated variants from non-breast tissue, and using expression quantitative trait loci, identify potential target genes of known breast cancer susceptibility variants, as well as 11 regions not previously known to be associated with breast cancer risk.

Introduction Physical activity may affect health via DNA methylation. The epigenetic influences of sedentary behaviors such as television viewing are unknown. We performed a genomewide study of DNA methylation in peripheral blood in relation to physical activity and television viewing time. Methods DNA methylation was measured using the Illumina Infinium HumanMethylation450K BeadChip array in blood samples collected at baseline (N = 5513) and follow-up (N = 1249) from participants in the Melbourne Collaborative Cohort Study. At baseline, times per week of leisure-time physical activity were self-reported. At follow-up, the International Physical Activity Questionnaire was used to assess MET-hours per week of total and leisure-time physical activity and hours per day of television viewing time. Linear mixed models were used to assess associations between physical activity and television viewing measures and DNA methylation at individual CpG sites, adjusted for potential confounders and batch effects. Results At follow-up, total physical activity was associated with DNA methylation at cg10266336 (P = 6.0 × 10−9), annotated to the SAA2 gene. Weaker evidence of associations (P < 1.0 × 10−5) were observed for an additional 14 CpG sites with total physical activity, for 7 CpG sites with leisure-time physical activity, and for 9 CpG sites with television viewing time. Changes in leisure-time physical activity between baseline and follow-up were associated with methylation changes (P < 0.05) at four of the seven CpG sites with weaker evidence of cross-sectional associations with leisure-time physical activity. Conclusion Physical activity and television viewing may be associated with blood DNA methylation, a potential pathway to chronic disease development. Further research using accelerometer data and larger sample sizes is warranted.

We examined the associations between germline variants and breast cancer mortality using a large meta-analysis of women of European ancestry. Meta-analyses included summary estimates based on Cox models of twelve datasets using ~10.4 million variants for 96,661 women with breast cancer and 7697 events (breast cancer-specific deaths). Oestrogen receptor (ER)-specific analyses were based on 64,171 ER-positive (4116) and 16,172 ER-negative (2125) patients. We evaluated the probability of a signal to be a true positive using the Bayesian false discovery probability (BFDP). We did not find any variant associated with breast cancer-specific mortality at P < 5 × 10−8. For ER-positive disease, the most significantly associated variant was chr7:rs4717568 (BFDP = 7%, P = 1.28 × 10−7, hazard ratio [HR] = 0.88, 95% confidence interval [CI] = 0.84–0.92); the closest gene is AUTS2. For ER-negative disease, the most significant variant was chr7:rs67918676 (BFDP = 11%, P = 1.38 × 10−7, HR = 1.27, 95% CI = 1.16–1.39); located within a long intergenic non-coding RNA gene (AC004009.3), close to the HOXA gene cluster. We uncovered germline variants on chromosome 7 at BFDP < 15% close to genes for which there is biological evidence related to breast cancer outcome. However, the paucity of variants associated with mortality at genome-wide significance underpins the challenge in providing genetic-based individualised prognostic information for breast cancer patients.

Background: DNA methylation may be one of the mechanisms by which alcohol consumption is associated with the risk of disease. We conducted a large-scale, cross-sectional, genome-wide DNA methylation association study of alcohol consumption and a longitudinal analysis of repeated measurements taken several years apart. Methods: Using the Illumina Infinium HumanMethylation450 BeadChip, DNA methylation measures were determined using baseline peripheral blood samples from 5,606 adult Melbourne Collaborative Cohort Study (MCCS) participants. For a subset of 1,088 of them, these measures were repeated using blood samples collected at follow-up, a median of 11 years later. Associations between alcohol intake and blood DNA methylation were assessed using linear mixed-effects regression models adjusted for batch effects and potential confounders. Independent data from the LOLIPOP (N=4,042) and KORA (N=1,662) cohorts were used to replicate associations discovered in the MCCS. Results: Cross-sectional analyses identified 1,414 CpGs associated with alcohol intake at P<10-7, 1,243 of which had not been reported previously. Of these 1,243 novel associations, 1,078 were replicated (P<0.05) using LOLIPOP and KORA data. Using the MCCS data, we also replicated (P<0.05) 403 of 518 associations that had been reported previously. Interaction analyses suggested that associations were stronger for women, non-smokers, and participants genetically predisposed to consume less alcohol. Of the 1,414 CpGs, 530 were differentially methylated (P<0.05) in former compared with current drinkers. Longitudinal associations between the change in alcohol intake and the change in methylation were observed for 513 of the 1,414 cross-sectional associations. Conclusion: Our study indicates that, for middle-aged and older adults, alcohol intake is associated with widespread changes in DNA methylation across the genome. Longitudinal analyses showed that the methylation status of alcohol-associated CpGs may change with changes in alcohol consumption.