Cigarette smoking and mammographic density in the Danish Diet, Cancer and Health cohort

Purpose Smoking before first childbirth increases breast
cancer risk, but the biological mechanism remains
unknown and may involve mammographic density (MD),
one of the strongest biomarkers of breast cancer risk. We
aimed to examine whether active smoking and passive
smoking were associated with MD.
Methods For the 5,356 women (4,489 postmenopausal)
from the Danish Diet, Cancer and Health cohort
(1993–1997) who attended mammographic screening in
Copenhagen (1993–2001), we used MD (mixed/dense or
fatty) assessed at the first screening after cohort entry.
Active smoking (status, duration, and intensity) and passive
smoking were assessed at cohort baseline (1993–1997) via
questionnaire, together with other breast cancer risk factors.
Logistic regression was used to estimate associations
(odds ratios, 95 % confidence intervals) between smoking
and MD, adjusting for confounders.
Results Two thousand and twenty-six (56.5 %) women
had mixed/dense MD, 2,214 (41.4 %) were current, and
1,175 (21.9 %) former smokers. Current smokers had
significantly lower odds (0.86, 0.75–0.99) of having mixed/
dense MD compared to never smokers, while former
smoking was not associated with MD. Inverse association
between smoking and MD was strongest in women who
initiated smoking before age of 16 years (0.79, 0.64–0.96),
smoked C15 cigarettes/day (0.83, 0.71–0.98), smoked C5
pack-years (0.62, 0.43–0.89), smoked [30 years (0.86,
0.75–0.99), and smoked C11 years before first childbirth
(0.70, 0.51–0.96). Association between smoking and MD
diminished after smoking cessation, with increased odds of
having mixed/dense breasts in women who quit smoking
[20 years ago as compared to current smokers (1.37,
1.01–1.67). There was no association between passive
smoking and MD.
Conclusions We found an inverse association between
active smoking and MD.