Women who have ever used the types of oral contraceptives that are common today, which contain low doses of estrogen and progestin, are 50% less likely than never-users of the pill to develop ovarian cancer; this level of protection is identical to that afforded by older oral contraceptives with higher hormone doses.1 These findings, from a population-based, case-control study, suggest that one of the main noncontraceptive benefits of the pill--its ability to prevent ovarian cancer--has been unaffected by changes made in pill formulations to make the method safer. Similarly, evidence from another case-control study suggests that ever-use of oral contraceptives reduces the odds of noncancerous ovarian tumors by 20%, and the decrease in risk is independent of estrogen dose.2 In both studies, the magnitude of protection increased with longer durations of pill use.
The ovarian cancer study was based on data for women aged 20-69 who were treated for epithelial ovarian cancer between 1994 and 1998 at 39 hospitals in Delaware, New Jersey and Pennsylvania. Eligible women had received their diagnosis within the previous six months, and all diagnoses were confirmed pathologically. The control group consisted of randomly selected women who were matched to ovarian cancer patients by five-year age-group and telephone exchange (for those younger than 65) or by county of residence (for 65-69-year-olds). Researchers interviewed all participants to gather information on their demographic characteristics, sexual and reproductive history, family cancer history and contraceptive use, including the brand of any oral contraceptive used. The analyses include interview data from 767 women with ovarian cancer and 1,367 controls.
To assess the effects of different hormone doses in combined oral contraceptives, the researchers classified pills as follows: Preparations containing less than 50 mcg of ethinyl estradiol or less than 100 mcg of mestranol were considered to have a low dose of estrogen; all others were considered high-dose estrogen pills. The progestin dose was considered low if the progestin was estimated to be less potent than 0.5 mg of norgestrel and was classified as high otherwise.
Some 80-90% of both cancer patients and controls were aged 40 or older, were white and had at least a high school education. Roughly three-quarters of cancer patients and nine in 10 controls had ever been pregnant. In all, 56% of women with ovarian cancer had ever used oral contraceptives, as had 69% of controls; the majority of pill users had taken pills containing low doses of both estrogen and progestin.
In analyses adjusting for age, race, pregnancy history and family history of ovarian cancer, women who had ever used oral contraceptives had a 40% lower risk of ovarian cancer than women who had never used the pill (odds ratio, 0.6). The protective effect increased with the duration of use: Women had a 30% decrease in risk (odds ratio, 0.7) if they had used the pill for up to four years and a 70% decrease (0.3) if they had used it for 10 or more years. Of note, the risk of ovarian cancer was reduced by 30% even after less than one year of oral contraceptive use, and it remained lowered for 30 years after termination of use.
The degree of protection against ovarian cancer was similar whether women began using the pill before 1972, when hormone doses were generally high (odds ratio, 0.7); between 1972 and 1980, when lower-dose formulations were being introduced (0.5); or after 1980, when low-dose pills dominated the market (0.6). Moreover, it varied little with different doses of estrogen and progestin. Women who used formulations with either high doses or low doses of both hormones experienced a 50% decrease in cancer risk (odds ratio, 0.5). Regimens with a high estrogen and low progestin dose provided a 30% decrease in risk (0.7), and low- estrogen, high-progestin pills lowered the risk by 40% (0.6).
When the data were further adjusted for duration of pill use, women still had similar reductions in ovarian cancer risk regardless of whether they used pills with high or low estrogen and progestin doses. Neither age at initiation of use nor duration since last use altered the protective effect of oral contraceptives.
The investigators note that their study is one of the first with a large group of patients and sufficiently long follow-up to evaluate the impact of low-dose oral contraceptives in comparison with that of high-dose formulations. Despite the difficulty in ascertaining pill brands many years after use, they conclude that low-dose preparations are as effective as high-dose oral contraceptives in preventing ovarian cancer.
Benign Ovarian Tumors
A total of 746 women aged 18-74 with surgically confirmed benign ovarian tumors were studied at six hospitals in the New York City area between 1992 and 1993. A group of 404 women, randomly selected from among those receiving gynecologic care at the same hospitals and frequency-matched to the subjects' age distribution, served as controls. Interviews were conducted to ascertain the women's background characteristics, reproductive history, personal and family medical histories and details about combined oral contraceptive use.
Women who had ever used oral contraceptives had a 20% lower risk of developing benign ovarian tumors than never-users of the pill (odds ratio, 0.8). Although a protective effect was not evident for women who reported current use of up to five years, the analysts found a strong trend of decreasing risk as duration of use increased; women who had been current users for more than five years were 60% less likely than never-users to develop benign ovarian tumors.
The pill's protective effect persisted even for women who had stopped using the method more than five years in the past. Again, the effect increased as duration of use rose and was most prominent for women who had used oral contraceptives for more than five years (odds ratio, 0.6).
Finally, the researchers assessed the impact of estrogen dose on the risk of noncancerous ovarian tumors. (They were unable to include progestin in the dosage analysis because the numbers of women who had taken pills with any particular progestin were small.) No differences in the tumor risk were found between women who had used high-dose pill formulations and those who had used oral contraceptives containing low doses of estrogen. --L. Ninger
1. Ness RB et al. and the SHARE Study Group, Risk of ovarian cancer in relation to estrogen and progestin dose and use characteristics of oral contraceptives, American Journal of Epidemiology, 2000, 152(3):233-241.
2. Westhoff C et al., Oral contraceptives and benign ovarian tumors, American Journal of Epidemiology, 2000, 152(3):242-246.