Meta-analysis: Use of combined oral contraceptives in the past 10 years is associated with an increased risk for breast cancer
ACP J Club. 1996 Nov-Dec;125:77. doi:10.7326/ACPJC-1996-125-3-077
Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet. 1996 Jun 22;347:1713-27. [PubMed ID: 8656904]
To determine whether an association exists between use of hormonal contraceptives and risk for breast cancer.
Studies were identified from review articles, computer-aided literature searches, and colleagues.
Studies were selected if they included at least 100 women with breast cancer and if information was obtained on the use of hormonal contraceptives and on reproductive history.
The principal investigators of all identified studies were invited to collaborate. If they agreed, they provided data on individual women pertaining to sociodemographic factors, use of hormonal contraceptives and hormone replacement therapy, family history of breast cancer, height, weight, age at menarche, reproductive history, menopausal status, age at menopause, gynecologic surgery, and tumor spread for those who had breast cancer.
54 studies involving 53 297 women with invasive breast cancer and 100 239 women without breast cancer provided data (90% of eligible studies). All analyses were stratified by study; age at diagnosis; parity; and, when appropriate, the age of the woman when her first child was born and the age at which she was no longer able to conceive. Overall, the relative risk (RR) for breast cancer in women who had used oral contraceptives compared with women who had never used them was 1.07 (P < 0.001). An increased risk for breast cancer existed in women who were currently using combined oral contraceptives (RR 1.24, 99% CI 1.15 to 1.33, P < 0.001), in those who had stopped using oral contraceptives 1 to 4 years previously (RR 1.16, CI 1.08 to 1.23, P < 0.001), and in those who had stopped using oral contraceptives 5 to 9 years previously (RR 1.07, CI 1.02 to 1.13, P = 0.009). No increased risk for breast cancer was found among women who had stopped using oral contraceptives for 10 or more years (RR 1.01, CI 0.96 to 1.05). The breast cancer diagnosed in women who had used oral contraceptives was less advanced than in those who had never used the contraceptives. For those who had used oral contraceptives compared with those who had not, the RR for tumors that spread beyond the breast compared with localized tumors was 0.88 (CI 0.81 to 0.95, P = 0.002). Duration of hormonal contraceptive use, age at first use, and the dose and type of hormone had little additional effect on the risk for breast cancer when recency of use was considered.
A small increased risk for breast cancer exists while women are taking combined oral contraceptives and in the 10 years after they stop. No evidence was found for an increased risk for breast cancer among women who stopped using oral contraceptives more than 10 years previously.
Source of funding: Imperial Cancer Research Fund.
For article reprint: Professor V. Beral, Collaborative Group on Hormonal Factors in Breast Cancer, ICRF Cancer Epidemiology Unit, Gibson Building, Radcliffe Infirmary, Oxford 0X2 6H3, England, UK. FAX 44-1865-310-545.
Concerns about the health effects of oral contraceptives have led to many studies that explore the effects on the risk for breast cancer. The results from most studies suggest a small risk from oral contraceptive use for younger women and no risk for older women.
As is usual for meta-analyses, the overall results do not substantially alter one's understanding of the previous studies and basically confirm a minimal, if any, increased risk for breast cancer. The real benefit of meta-analysis is in exploring subgroups. Here the results of the analysis indicate that the increased risk for breast cancer is limited to the time interval during which oral contraceptives were used and shortly thereafter, with no long-term effects, and that the excess number of cases of breast cancer seen are generally local disease.
It is important to consider the absolute risk associated with oral contraceptives. In particular, given the relatively low incidence of breast cancer among women in the age groups who most often use oral contraceptives, an excess risk of 20% to 30% still seems inconsequential, particularly if these breast cancers are at a very early stage (1). Certainly, the excess risk in terms of breast cancer incidence seems minuscule when compared with the potential hazards from cardiovascular or other adverse health events. For policy-making purposes, the effect of oral contraceptive use on breast cancer mortality would be useful.
Questions have recently been raised about whether epidemiologic methods can resolve concerns about small RRs (< 2) through use of observational studies. Meta-analyses, such as this one, are probably as good as we can do and at least suggest that there is not a greater cause for concern. Nonetheless, this study reminds me of what a professor of mine once said: We usually say that the difference was small but statistically significant. Perhaps we should say that the difference was statistically significant but small.
Alfred I. Neugut, MD, PhD
Columbia UniversityNew York, New York, USA