Oral contraceptives and breast, endometrial, and ovarian cancers. The Cancer and Steroid Hormone Study Group, Atlanta Georgia.

Abstract

The Cancer and Steroid Hormone Study was designed to minimize certain problems criticized in some previous case-control studies of oral contraceptives (OCs) and gynecological cancer. The largest of its type undertaken thus far, the study included over 4000 women with breast cancer, over 600 with endometrial cancer, and over 500 with ovarian cancer, along with over 4000 controls. The controls were selected from the population in the geographical areas from which the cases were drawn. The study was limited to women 20-54 years. Study enrollment occurred between December 1980 and December 1982 at 8 regional cancer registries participating in the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute (NCI). Investigators at these centers attempted to identify all women 20-54 years of age with newly diagnosed primary breast, endometrial, or ovarian cancer. These women constituted the case group. Controls were women 20-54 years chosen by an established method of telephoning households at random in the geographical regions covered by the participating centers. All controls were interviewed in person. To improve the accuracy of reporting OC and other hormone use, 2 memory aids were used: use of a book of photographs of all OC and other female hormone preparations marketed in the US up to 1980; and use of a calendar to record information for each woman about such major life events as marriages, births, divorces, and deaths. Information on contraception and other reductive events was then recorded around these usually well-remembered milestones. A panel of pathologists reviewed histological specimens for all cases of endometrial and ovarian cancer enrolled in this study and for all cases and controls with a history of benign breast disease for whom a biopsy specimen was available. The major findings with respect to OC use and risk of cancers of the breast, endometrium, and ovary largely confirmed those in previous studies. It was found that a history of OC use appeared to protect against endometrial and ovarian cancer and had no overall effect on the risk of breast cancer. Women who had used OCs for less than 1 year had the same risk of developing endometrial cancer as "never-users." Compared with never-users, women who used OCs for 1-5 years, or for more than 5 years, had a relative risk of 0.4. Women who have ever used combination OCs had about 1/2 the risk of developing ovarian cancer compared to women who have never used them. Compared with women who never used OCs, women who had ever used them had a relative risk of breast cancer of 0.9. There was no appreciable trend in the relative risk of breast cancer by duration of use.