Skip to content

Breast Cancer Health Center

Font Size

Breast Cancer Prevention (PDQ®): Prevention - Health Professional Information [NCI] - Description of Evidence

Incidence of Outcomes Per 1,000 Women continued...

These findings are similar to those for women who undergo castration for nononcologic diagnoses. Women treated with thoracic radiation who undergo radiation therapy or chemotherapy, which often results in ovarian ablation, also have similar findings.


A multicenter phase III RCT of fenretinide versus no treatment was performed in 2,867 women who received local therapy for stage 0 (DCIS) or stage I (T1–T2, N0, M0; T = tumor, N = node, M = metastasis) breast cancer. An analysis at 8 years showed no difference in contralateral or ipsilateral breast cancer, but a post hoc analysis revealed differential effects for premenopausal and postmenopausal women.[114] A subsequent analysis at 15 years of the 1,739 women enrolled at the organizing center confirmed the beneficial effect in premenopausal women, reducing both contralateral and ipsilateral cancers, HR = 0.62 (95% CI, 0.46–0.83). This beneficial effect was age dependent, with the youngest women achieving the most benefit.[115] Although the daily fenretinide 200 mg was withheld for 3 days each month, there was a cumulative incidence of low-grade dark adaptation (i.e., night blindness) and dermatologic disorders. As with any vitamin A analog, women taking this drug should avoid pregnancy because of potential teratogenic effects.

Factors of Unproven or Disproven Association


Abortion has been suggested as a cause of subsequent breast cancer. Studies showing an association used recalled information in populations in which induced abortion had a social or religious stigma, differential reporting of prior abortion by breast cancer patients, and controls. Trials conducted in social environments where abortion is accepted, however, have not shown an association with breast cancer.[116,117,118,119,120,121]

A meta-analysis of women from 53 studies in 16 countries with liberal abortion laws was performed.[122] Analyses were performed separately on 44,000 women with breast cancer who had information on abortion collected prospectively (i.e., 13 studies) versus 39,000 women with breast cancer from whom information was collected retrospectively (i.e., 40 studies). The RR of breast cancer for women with spontaneous abortion was 0.98 (95% CI, 0.92–1.04 for those with prospective data collection and 0.94–1.02 for retrospective data). The RR after induced abortion was 0.93 (95% CI, 0.89–0.96; P = .0002) if the information was collected prospectively but was 1.11 (95% CI, 1.06–1.16) if it was collected retrospectively. Additional analyses of the number and timing of aborted pregnancies were performed, but none showed a significant association with breast cancer.[122]

Oral contraceptives

Oral contraceptives have been associated with a small increased risk of breast cancer in current users that diminishes over time.[123] A well-conducted case-control study did not observe an association between breast cancer risk and oral contraceptive use for every use, duration of use, or recency of use.[124]

Next Article:

Today on WebMD

Breast Cancer Overview
From self-exams and biopsies to reconstruction, we’ve got you covered.
Dealing with breast cancer
Get answers to your questions.
woman having mammogram
Experts don’t agree on all fronts, but you can be your own advocate.
woman undergoing breast cancer test
Many women worry. But the truth? Most abnormalities aren’t breast cancer.
Breast Cancer Treatments Improving
Resolved To Quit Smoking
Woman getting mammogram
Screening Tests for Women
ovarian cancer overview slideshow
serious woman
what is your cancer risk
10 Ways to Revitalize Slideshow