Abortion and breast cancer risk for Australian women

To the Editor: Recent comments in the media have renewed attention on the question of whether abortion is a risk factor for breast cancer, despite broad acceptance internationally in the scientific, medical and cancer advocacy communities that there is no association.

It was demonstrated in 2004 that abortion is not a risk factor for breast cancer, based on a combined analysis of prospective cohort study data from around the world.1 The authors showed empirically that retrospective case–control studies tend to give biased findings, most likely because women with breast cancer (cases) are more likely than controls to report previous abortions. Prospective cohort studies avoid this bias because information on abortions is collected before any of the women develop the disease. Indeed, a recent meta-analysis of findings from studies of Chinese women reported an apparent increased breast cancer risk associated with abortion based on case–control studies, but no evidence of association based on prospective cohort studies.2 Several other large prospective cohort studies have published consistently null results since 2004.3,4

In 1990–1994, the Melbourne Collaborative Cohort Study5 recruited 24 018 Victorian women aged 40–69 years who were free of breast cancer, and collected information on risk factors, including all pregnancies. The questionnaire did not differentiate between spontaneous and induced abortions. Using record linkage with the Victorian Cancer Registry, we determined that 1235 of these women were subsequently diagnosed with breast cancer, up to 30 December 2012. We assessed associations prospectively using Cox regression, with age as the time variable, considered from recruitment to breast cancer diagnosis, and censored at death, permanent departure from Victoria or 30 December 2012. Analyses were adjusted for country, year of birth and socioeconomic status.

Ever having had an abortion in the first trimester was not associated with breast cancer risk (hazard ratio [HR], 1.06; 95% CI, 0.94–1.20). The result was similar after additional adjustment for age at menarche, oral contraceptive use, number of full-term pregnancies, age at first birth and lifetime breastfeeding (HR, 1.07; 95% CI, 0.94–1.22). The result was also similar for abortion before first full-term pregnancy (HR, 0.97; 95% CI, 0.81–1.16) and for abortion in the first two trimesters (HR, 1.04; 95% CI, 0.92–1.17).

Consistent with international research findings, abortion is not associated with risk of breast cancer for Australian women.