Does exposure to smoke increase risk of breast cancer?
Clinical bottom line
This study finds that active smoking increases breast cancer risk and when women stop smoking the risk disappears.
Background
Despite considerable research, the relationship between smoking and breast cancer is still controversial. This prospective study examines the association between breast cancer risk and both active and passive smoking. It aims to examine smoking history and genetic predisposition, which could be responsible for the inconsistent results.
Reference
P Reynolds et al. Active smoking, household passive smoking, and breast cancer: evidence from the California Teachers Study. Journal of the National Cancer Institute 2004 96: 29-37.
Study
Participants were 116,544 women from the 1995 California Teachers Study, with no previous breast cancer diagnosis. They were followed for five years. Through mailed questionnaires the following information was obtained:
- Smoking status: never, former or current.
- Current smokers: average number of cigarettes smoked per day; years of smoking; smoking pack-years (i.e. number of packs smoked per day times number of years smoked); age at which smoking started; smoking behaviour in relation to first full-term pregnancy.
- Former smokers: number of years since quitting.
- Passive smokers: defined as lifetime never smokers exposed to household passive smoking (categorised as childhood exposure only, adulthood exposure only or both).
- Personal risk factors: family history of breast cancer (defined as breast cancer in a first-degree relative), age, ethnicity, body mass index (BMI), physical activity, alcohol consumption, age at menarche, pregnancy history, menopausal status, BMI and menopausal status interaction and hormone therapy use.
Results
The majority of women were non-Hispanic white (87%) and approximately half were post-menopausal.Table 1 shows the distribution of smoking status among the women with number of breast cancer cases identified. Results were adjusted for these personal risk factors.
Table 1. Distribution of smoking status among cohort with incidence of breast cancer
| Never smoked | ||
| Passive exposure | ||
| Former smokers | ||
| Current smokers | ||
| Total |
Passive smoking
No associations were found between breast cancer risk and exposure to household passive smoking (childhood exposure only: hazard ratio 0.92, 95% confidence interval 0.78 to 1.07; adult exposure only: hazard ratio 0.94, 95% confidence interval 0.79 to 1.12; childhood and adulthood exposure: hazard ratio 0.93, 95% confidence interval 0.79 to 1.09).
Current and former smoking
The analysis examining breast cancer risk among current and former smokers was run twice: (1) relative to never smokers, including passive smokers; and (2) relative to never smokers, excluding passive exposure.
Current smoking. Regardless of the reference group used, current smoking was associated with an increased risk of breast cancer (hazard ratio 1.32, 95% confidence interval 1.10 to 1.57, relative to all never smokers; and hazard ratio 1.25, 95% confidence interval 1.02 to 1.53, relative to never smokers excluding passive exposure).
Former smoking. Breast cancer risk was not associated with (i) former smoking (hazard ratio 1.08, 95% confidence interval 0.98 to 1.19, relative to all never smokers; and hazard ratio 1.03, 95% confidence interval 0.89 to 1.18, relative to never smokers excluding passive exposure) or (ii) the number of years since quitting.
Current smoking - further analysis. Years of smoking and smoking in relation to first full-term pregnancy were not related to breast cancer risk. Relative to all never smokers, an increased risk was found:
- in women averaging 20 or more cigarettes per day (hazard ratio 1.22, confidence interval 1.05 to 1.42);
- when the number of smoking pack-years was 31 or more (hazard ratio 1.25, confidence interval 1.06 to 1.47);
- when women started smoking before the age of 20 (hazard ratio 1.17, confidence interval 1.05 to 1.30).
Comment
This study examined several factors associated with smoking history, which sometimes resulted in small sample sizes. With larger samples, more associations may have been found.
This paper analysed risk associated with a family history of breast cancer. Smoking was associated with an increased risk among women without a history of breast cancer, but not among women with such a history. These results were not reported here because there were only 26 cases in the latter analysis.
The further analysis of current smokers included both current and former smokers. The addition of former smokers may have been to boost sample sizes, but may have confounded the results (having observed that former smoking was not related to breast cancer risk).
An increased risk of breast cancer was found among active smokers, but not passive smokers. Note that this study only looked at household passive smoking; it did not include exposure at work or in social settings and did not include measures of intensity or duration. The larger sample here overturns a positive association found in a previous but much smaller study.
Bandolier (Is there is a link between passive smoking and breast cancer?) raises the possibility that a woman's risk of breast cancer increases with exposure to cigarette smoke at a young age (during times of rapid breast tissue development). In this paper, women who started to smoke before the age of 20 had an increased risk, which supports this theory (although smoking at an early age might be correlated with smoking for a long time as well as with smoking before a first pregnancy).
Alcohol consumption could confound associations between smoking and breast cancer. However, when this paper restricted its analysis to 35,123 non-drinkers (but with an unknown number of cases) current smokers continued to have an increased risk of breast cancer (hazard ratio of 1.66, 95% confidence interval 1.15 to 2.40).
Good news for former smokers! Breast cancer risk was not related to former smoking or the number of years since quitting, so it appears that once a smoker has given up, there is no longer a risk of breast cancer.
Fear of breast cancer may provide the catalyst that some women need to stop smoking. For teenagers and young women, the increased risk of breast cancer could be a particularly effective deterrent not to smoke.