Estrogen Replacement Therapy and Risk of Ovarian Cancer

By the middle of the 20th century, it was well recognized that elderly women frequently developed severe osteoporosis, resulting in a life complicated by constant back pain and repeated fractures. By the 1970s and 1980s, it became clear that use of estrogenic substances at or near the time of menopause could prevent or treat osteoporosis, and these drugs became widely prescribed and taken. Even before the bone-sparing effects of estrogen were known, these agents were used extensively for the treatment of menopausal symptoms, primarily vasomotor instability and vaginal atrophy.¹

An intriguing coincidence occurred from about 1955 to 1965. Reproductive-aged women from that generation were the first to experience substantially reduced pregnancy risks due to development of safe anesthesia and readily available transfusion, as well as the availability of antihypertensive agents and broad-spectrum antibiotics. During this time, effective oral contraception became widely available to the general population at a reasonable cost. The ability to prevent pregnancy effectively further reduced reproductive-associated risk. By the 1980s and 1990s, as a result of these and other medical advances, the US population of postmenopausal women began to increase dramatically. At the same time, the incidence of cardiovascular, neoplastic, and neurologic diseases among older persons began to soar.

For a short time, estrogen replacement was viewed as the perfect solution for many health problems in postmenopausal women. Estrogens were thought to prevent coronary artery disease² and delay the onset of Alzheimer disease.^{3 - 4} The benefits of preserving bone⁵ and reducing menopausal symptoms were already well-known.⁶ It is not surprising that the pharmaceutical industry developed numerous estrogenic agents, which could be swallowed, placed in the vagina, applied with a patch, or rubbed into the skin, and that these agents have been prescribed for and are now used by millions of women. However, recent secondary prevention studies provide compelling evidence that there is no protection against further cardiovascular events for women with coronary heart disease who take hormone replacement therapy (HRT) and there may be real harm.^{7 - 8}

In this issue of THE JOURNAL, Lacey et al⁹ report the results of their follow-up study of a large cohort of women who were recruited in the 1970s to participate in the Breast Cancer Detection Demonstration Project. In 1979, the National Cancer Institute added questions concerning ovarian cancer and its suspected risk factors (including HRT) to follow-up questionnaires. The analyses of these data by Lacey et al show that women who used estrogen-only HRT had a significantly increased risk of later development of ovarian cancer. The fact that the association increased with longer duration of therapy, particularly with duration of use for 10 years or longer, increases the face validity of their findings.

The study by Lacey et al is not the first to examine the association between HRT and ovarian cancer. Most early studies found no association, although more recent studies have reported consistent increases in the risk ratio for ovarian cancer among women who have taken estrogen-only HRT compared with those who have not.^{10 - 12} While the data from these observational studies do not establish causality, the association between estrogen use and ovarian cancer should be worrisome enough for clinicians to consider carefully whether to suggest estrogen-only HRT.

While it is still common for women who have had a hysterectomy (but who retain 1 or both ovaries) to be given estrogen only, most women who did not have a hysterectomy and who receive HRT usually are prescribed an estrogen-progestin combination to reduce the risk of endometrial cancer.^{13 - 14} Few studies have found an association between ovarian cancer and use of combined estrogen-progestin agents. Riman et al¹² identified an increased risk of ovarian cancer among women who took estrogen plus sequential progestins, but not among those who used continuous combination estrogen-progestin regimens. In the study by Lacey et al, the risk of ovarian cancer was not increased among women who used short-term–estrogen-progestin regimens, although these findings were based on a small number of women. Since it has been common practice to prescribe both drugs only in recent years, it is possible that insufficient time has passed for a valid examination of combined HRT and ovarian cancer risk.

Estrogen replacement therapy certainly is not the panacea it once appeared. Physicians counseling women about HRT must consider the unique needs of each patient, and attempt to weigh the benefits and risks on an individual basis. At present, certain benefits such as prevention of osteoporosis and treatment of vasomotor symptoms are known. In assessing risk, postmenopausal women who have not had a hysterectomy should not receive estrogen only, as there is an unquestioned increased risk of endometrial cancer, and perhaps an increased risk of breast cancer.¹⁵ Based on the information in the study by Lacey et al, these risks now include ovarian cancer.