• Epidemiology Research Program, American Cancer Society, Atlanta, GA, USA.
• Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD, USA.
• Department of Community Medicine, Faculty of Health Sciences, UiT The Arctic University of Norway, Tromsø, Norway.
• Department of Epidemiology and Biostatistics, School of Public Health, Indiana University, Bloomington, IN, USA.
• Cancer Epidemiology Centre, Cancer Council Victoria, Melbourne, VIC, Australia and Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Victoria, Australia.
• Epidemiology Branch, National Institute of Environmental Health Sciences, NIH/DHHS, Research Triangle Park, NC, USA.
• Department of Epidemiology and Biostatistics, Milken Institute School of Public Health, George Washington University, Washington, DC, USA.
• School of Public Health, University of Minnesota, Minneapolis, MN, USA.
• Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA.
• Department of Oncology, School of Medicine, Georgetown University, Washington, DC, USA.

BACKGROUND: The 2014 US Surgeon General's report noted research gaps necessary to determine a causal relationship between active cigarette smoking and invasive breast cancer risk, including the role of alcohol consumption, timing of exposure, modification by menopausal status and heterogeneity by oestrogen receptor (ER) status.

METHODS: To address these issues, we pooled data from 14 cohort studies contributing 934 681 participants (36 060 invasive breast cancer cases). Cox proportional hazard regression models were used to calculate multivariable-adjusted hazard ratios (HRs) and 95% confidence intervals (CIs).

RESULTS: Smoking duration before first birth was positively associated with risk ( P -value for trend = 2 × 10 -7 ) with the highest HR for initiation >10 years before first birth (HR = 1.18, CI 1.12-1.24). Effect modification by current alcohol consumption was evident for the association with smoking duration before first birth ( P -value=2×10 -4 ); compared with never-smoking non-drinkers, initiation >10 years before first birth was associated with risk in every category of alcohol intake, including non-drinkers (HR = 1.15, CI 1.04-1.28) and those who consumed at least three drinks per day (1.85, 1.55-2.21). Associations with smoking before first birth were limited to risk of ER+ breast cancer ( P -value for homogeneity=3×10 -3 ). Other smoking timing and duration characteristics were associated with risk even after controlling for alcohol, but were not associated with risk in non-drinkers. Effect modification by menopause was not evident.

CONCLUSIONS: Smoking, particularly if initiated before first birth, was modestly associated with ER+ breast cancer risk that was not confounded by amount of adult alcohol intake. Possible links with breast cancer provide additional motivation for young women to not initiate smoking.