Select Page

Mayo Clin Proc. 2006;81:1290-1302

Mayo Foundation for Medical Education and Research

Oral Contraceptive Use as a Risk Factor for Premenopausal Breast Cancer: A Meta-analysis

OBJECTIVE: To perform a meta-analysis of case-control studies that addressed whether prior oral contraceptive (OC) use is associated with premenopausal breast cancer.




METHODS: We searched the MEDLINE and PubMed databases and bibliography reviews to identify case-control studies of OCs and premenopausal breast cancer published in or after 1980. Search terms used included breast neoplasms, oral contraceptives, contraceptive agents, and case-control studies. Studies reported in all languages were included…

RESULTS: Use of OCs was associated with an increased risk of premenopausal breast cancer in general (OR, 1.19; 95% CI, 1.09-1.29) and across various patterns of OC use. Among studies that provided data on nulliparous and parous women separately, OC use was associated with breast cancer risk in both parous (OR, 1.29; 95% CI, 1.20-1.40) and nulliparous (OR, 1.24; 95% CI, 0.92-1.67) women. Longer duration of use did not substantially alter risk in nulliparous women (OR, 1.29; 95% CI, 0.85-1.96). Among parous women, the association was stronger when OCs were used before first full-term pregnancy (FFTP) (OR, 1.44; 95% CI, 1.28-1.62) than after FFTP (OR, 1.15; 95% CI, 1.06-1.26). The association between OC use and breast cancer risk was greatest for parous women who used OCs 4 or more years before FFTP (OR, 1.52; 95% CI, 1.26-1.82).

CONCLUSION: Use of OCs is associated with an increased risk of premenopausal breast cancer, especially with use before FFTP in parous women.

Mayo Clin Proc. 2006;81(10):1290-1302
CI=confidence interval; FFTP=first full-term pregnancy; OC=oral contraceptive; OR=odds ratio
Mayo Clin Proc. 2006;81:1287-1289

Mayo Foundation for Medical Education and Research

Oral Contraceptive Use and Breast Cancer Risk: Current Status
James R. Cerhan, MD, PhD
Division of Epidemiology Mayo Clinic College of Medicine Rochester, Minn

One might have thought that the issue of whether oral contraceptives (OCs) are associated with breast cancer risk would have been settled by now, given that these agents were introduced in the early 1960s and more than 60 case-control and 10 cohort studies, several meta-analyses,1-4 a very large pooled analysis,5 and a major monograph6 have addressed this issue.

On the basis of the accumulated data, the International Agency for Research on Cancer (IARC) classified oral estrogen-progestogen contraceptives as carcinogenic to humans (group 1 carcinogen) in 2005, which is a higher classification than the 1999 IARC evaluation.7

Of course things have changed, including OC formulations, the epidemiology of breast cancer, and patterns of use of OCs in the population. For example, the dosage of estrogen has decreased, new progestins (eg, desogestrel and norgestimate) have been introduced, the hormone-free interval has been shortened, and new delivery systems have become available.8

In the United States, breast cancer screening increased dramatically in the 1980s, breast cancer incidence rates increased in both the 1980s and the 1990s, and breast cancer mortality rates began declining in the early 1990s.9 In terms of use, a greater number of younger women are using OCs, and they are using them with increasing duration.10 Finally, our understanding of breast cancer pathogenesis has evolved.

Animal studies, mathematical modeling, migrant studies, and accumulating results from epidemiological studies suggest that breast tissue is particularly susceptible to carcinogenic insults before differentiation during the first full-term pregnancy.11,12

See also page 1290

In this context, the meta-analysis by Kahlenborn et al13 in the current issue of Mayo Clinic Proceedings allows us to take stock of the current state of our knowledge of OC use and premenopausal breast cancer risk, particularly in relationship to first full-term pregnancy.

They conducted a meta-analysis of 39 independent case-control studies that had most cases diagnosed since 1980 to better evaluate the association with breast cancer in the context of more contemporary use patterns.

Overall, they found that compared to never use, ever use of OCs was associated with a small but statistically significant increased risk of breast cancer (odds ratio [OR], 1.19; 95% confidence interval [CI], 1.09-1.29), although significant heterogeneity across individual studies could not be readily explained.

These results are generally comparable to the Oxford pooled analysis published in 1996,5 which was based on pooled original data from 54 epidemiological studies (case-control and cohort) representing 53,297 women with breast cancer and 100,239 control patients and was thought to represent about 90% of the epidemiological information on the topic available at that time. In that analysis, a weaker overall risk of breast cancer was found for ever use compared with never OC use (OR, 1.07).

The difference in risk between these 2 analyses is likely due in part to the studies included. In the Oxford pooled analysis, 66%

of cases were 45 years or older at the time of diagnosis, and 50% of the cases were diagnosed before 1984. In contrast, the current meta-analysis included only premenopausal women (or women <50 years), women who were diagnosed as having breast cancer mainly after 1980, and an additional 6 studies whose recruitment period ended in 1996 or later.

Of note, when the results of the Oxford pooled analysis were stratified by menopausal status, evidence showed that risk was somewhat greater for premenopausal (OR, 1.22) vs post-menopausal (OR, 1.08) women for recent (<5 years) OC users, although this did not hold for more distant former use.5

Another major finding from the current meta-analysis was that compared with never use, ever use of OCs before first full-term pregnancy was more strongly associated with breast cancer risk (OR, 1.44; 95% CI, 1.28-1.62) than ever use after first full-term pregnancy (OR, 1.15; 95% CI, 1.06-1.26) and that risk was even stronger for OC use for 4 or more years before first full-term pregnancy (OR, 1.52; 95% CI, 1.26-1.82).

A higher risk of breast cancer for OC use before first full-term pregnancy was first described more than 25 years ago by Pike et al,14 and subsequent meta-analyses have reported ORs in the range of 1.4 to 1.7 for this association.1-3 The Oxford pooled analysis found that compared with never OC users, current users who began use before first full-term pregnancy had a higher breast cancer risk (OR, 1.33) than current users who began OC use after that time (OR, 1.21). However, this effect decreased with time since last use, so that last use of OCs of 10 or more years was no longer associated with risk for either group.5 The current meta-analysis was unable to assess recency of last use, so the results cannot be directly compared.

The more comprehensive findings from the Oxford analysis highlight an important methodologic strength of the pooled analysis of original data in that study compared to a meta-analysis of published data in the current study. A meta-analysis generally relies on published estimates of effect and often cannot conduct more detailed analyses based on characteristics of use, adjust for confounding factors, or evaluate interactions. A pooled analysis also overcomes many statistical concerns regarding a meta-analysis of observational studies.15

Thus, for example, the Oxford pooled analysis5 was able to show that, while current users of OCs had an increased risk of breast cancer (OR, 1.24; 95% CI, 1.15-1.33), this risk decreased with time since last use so that at 10 or more years since last use there was no increase in risk compared with never users (OR, 1.01; 95% CI, 0.96-1.05). Duration of use, age at first use, and dose and type of hormone had little effect on breast cancer risk once recency of use was taken into account, and breast tumors diagnosed in women using OCs were more likely to be clinically localized compared with tumors in women who had never used OCs. Family history, weight, ethnicity, menopausal status, and alcohol use did not appear to interact with OC use to alter the risk of breast cancer.

Thus, while the meta-analysis by Kahlenborn et al raises the issue of OC use before first full-term pregnancy, a pooled analysis of individual level data from recent studies is needed to provide more definitive evidence. New studies may also need to be conducted to fully evaluate more recent OC use patterns associated with breast cancer risk.

How should clinicians counsel their patients at this time? First, OCs are extremely effective in preventing pregnancy when used correctly.8 Second, although OCs appear to be carcinogenic,7 the relative risk is small, and the absolute risk (excess breast cancers due to OC exposure) is very small. For example, the Oxford pooled analysis estimates that the excess number of cases of breast cancer expected to be diagnosed up to 10 years after discontinuation of OC use among 10,000 European or North American women is 0.5 cases for OC use from age 16 to 19 years, 1.5 cases for OC use from age 20 to 24 years, and 4.7 cases for OC use from 25 to 29 years. These cases are also likely to be clinically localized. Third, although a formal risk-benefit analysis is beyond the scope of this editorial, all risks and benefits of OC use must be considered, not just the risk of breast cancer. Other cancer risks may include cervical cancer and liver cancer in populations at low risk for hepatitis B viral infection. Additionally, IARC has determined that there is convincing evidence that OCs decrease the risk of ovarian and endometrial cancer, and there is accumulating evidence that they may lower the risk of colorectal cancer.7

Other major noncancer risks of OC use include ischemic stroke, venous thromboembolism, and myocardial infarction, but because these are rare events in women of childbearing age, the attributable risks are very small.8,16 Finally, there is a growing number of noncontraceptive health benefits associated with OCs, including relief from menstrual disorders; reduced risk of pelvic inflammatory disease, benign breast disease, uterine leiomyomas, and ovarian cysts; and improved bone mineral density.8

From the perspective of epidemiology and public health, we must continue to closely follow the epidemiology of OC use and health outcomes, given the widespread use of these agents and their high potential to impact women’s health in both a beneficial and a deleterious manner.

The current study highlights the need for a close evaluation of OC use before first full-term pregnancy since this is an important biologic issue with clear clinical and public health implications. Any association would also add additional support for identifying other exposures during the time before first full-term pregnancy associated with breast cancer risk in later life because identification of modifiable factors in this period would support expanding the window for breast cancer prevention to earlier in life.


Romieu I, Berlin JA, Colditz G.  Oral contraceptives and breast cancer: review and meta-analysis. Cancer. 1990;66:2253-2263.

Thomas DB.  Oral contraceptives and breast cancer: review of the epidemiologic literature. Contraception. 1991;43:597-642.

Delgado-Rodriguez M, Sillero-Arenas M, Rodriguez-Contreras R, Lopez Gigosos R, Galvez Vargas R.  Oral contraceptives and breast cancer: a meta-analysis. Rev Epidemiol Sante Publique. 1991;39:165-181.

Rushton L, Jones DR.  Oral contraceptive use and breast cancer risk: a meta-analysis of variations with age at diagnosis, parity and total duration of oral contraceptive use. Br J Obstet Gynaecol. 1992;99:239-246.


orative 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;347:1713-1727.

International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Vol 72: Hormonal Contraception and Post-menopausal Hormonal Therapy. Lyon, France: IARC; 1999.

Cogliano V, Grosse Y, Baan R, Straif K, Secretan B, El Ghissassi F.  WHO International Agency for Research on Cancer. Carcinogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment. Lancet Oncol. 2005;6:552-553.

Burkman RT.  Oral contraceptives: current status. Clin Obstet Gynecol. 2001;44:62-72.

Ries LAG, Harkins D, Krapcho M, et al. SEER Cancer Statistics Review, 1975-2003. Bethesda, Md: National Cancer Institute; 2006.

Marchbanks PA, McDonald JA, Wilson HG, et al. Oral contraceptives and the risk of breast cancer. N Engl J Med. 2002;346:2025-2032.

Colditz GA, Frazier AL.  Models of breast cancer show that risk is set by events of early life: prevention efforts must shift focus. Cancer Epidemiol Biomarkers Prev. 1995;4:567-571.

Russo J, Hu YF, Silva ID, Russo IH.  Cancer risk related to mammary gland structure and development. Microsc Res Tech. 2001;52:204-223.

Kahlenborn C, Modugno F, Potter DM, Severs WB.  Oral contraceptive use as a risk factor for premenopausal breast cancer: a meta-analysis. Mayo Clin Proc. 2006;81:1290-1302.

Pike MC, Henderson BE, Casagrande JT, Rosario I, Gray GE.  Oral contraceptive use and early abortion as risk factors for breast cancer in young women. Br J Cancer. 1981;43:72-76.

Blettner M, Sauerbrei W, Schlehofer B, Scheuchenpflug T, Friedenreich C.  Traditional reviews, meta-analyses and pooled analyses in epidemiology. Int J Epidemiol. 1999;28:1-9.

Pymar HC, Creinin MD.  The risks of oral contraceptive pills. Semin Reprod Med. 2001;19:305-312.

Analysis Suggests Small Increase in Risk When Oral Contraceptives Used Before First Pregnancy

Women who take oral contraceptives have a slightly increased risk of developing breast cancer early in life, with the risk being greatest for women who use oral birth control before they have their first child, a new analysis suggests.

Researchers combined the findings from 34 previous studies designed to examine the impact of oral contraceptive use on breast cancercancer diagnosed before menopausemenopause.

Twenty-one of 23 studies that followed women who took oral birth control before having their first child showed an increased risk of early breast cancer.

Measuring Risk

Based on these studies, the researchers concluded that taking oral contraceptives before a first full-term pregnancypregnancy increases premenopausal breast cancer risk by 44%, compared with women who have never used oral contraceptives.

The increase in risk was 52% among women who took the pill for four years or more before having their first child.

"I think women should know about this risk, and they are not being told," researcher Chris Kahlenborn, MD, of Altoona Hospital in Altoona, Pa., tells WebMD.

"Anyone who is prescribing oral contraceptives has a duty to tell women that 21 out of 23 studies showed an increased risk."

Breast cancer is most often diagnosed in women over the age of 50, but cancers that occur in younger women tend to be more aggressive. It is the leading cancer killer among women between the ages of 20 and 59 in the U.S.

Kahlenborn says he conducted the analysis because he believes far too few women know the risks associated with oral contraceptive use.

"As I studied the medical literature, I noticed that a trend appeared," he says. "Namely, oral contraceptive use prior to first full-term pregnancy seemed to consistently increase the risk of premenopausal breast cancer. Although the trend was apparent, premenopausal women have continued to hear that oral contraceptives are safe."

The 34 studies chosen for the analysis included women who were premenopausal or younger than 50 whose breast cancers had been diagnosed during or after 1980.

Studies examining breast cancers diagnosed before this were excluded in an attempt to approximate the risk with oral contraceptives as they are currently prescribed.

When all the studies were combined, use of oral birth control was associated with a 19% overall increased risk of breast cancer diagnosed before menopause. But the increase in risk more than doubled among women who took oral contraceptives before a first pregnancy.

Risk Still Small

In an editorial accompanying the study, Mayo Clinic epidemiologist James Cerhan, MD, PhD, points out that though the link between oral contraceptive use and early breast cancer appears real, the risk is still quite small.

Cerhan notes that a previous analysis calculated the risk of excess breast cancers occurring up to 10 years after stopping oral birth control. Researchers concluded that one excess breast canc

er could be expected for every 20,000 women who used oral contraceptives from ages 16 to 19 and 4.7 cancers could be expected for every 10,000 women who used them from ages 25 to 29.

According to Cerhan's editorial, there is evidence that oral contraceptive use can reduce a woman’s risk for ovarian and endometrial cancers, and recent studies suggest that it may also protect against colorectal cancercolorectal cancer.

He writes that use of oral birth control is also believed to reduce the risk of pelvic inflammatory disease, benign breast disease, ovarian cystsovarian cysts, and other reproductive-related health problems.

Cerhan concludes that if oral contraceptive use early in life increases a woman’s breast cancer risk, other early-life influences probably do, too.

SOURCES: Kahlenborn, C. Mayo Clinic Proceedings, October 2006; vol 81: pp 1290-1302. Chris Kahlenborn, MD, department of internal medicine, Altoona Hospital, Altoona, Pa. James R. Cerhan, MD, PhD, division of epidemiology, Mayo Clinic College of Medicine, Rochester, Minn. The Lancet, 1996; vol 347: pp 1713-1727.

[31October2006, The Pill May Raise Breast Cancer Risk, Salynn Boyles, WebMD Medical News  Reviewed By Louise Chang, MD]