Estrogen-Only Therapy After Menopause Is Linked to Ovarian Cancer

First published online:

| DOI: https://doi.org/10.1363/3431902b

Postmenopausal women who use estrogen-only replacement therapy are at increased risk for ovarian cancer, especially after 10 or more years of use.1 Postmenopausal women in a prospective cohort study who had ever used estrogen-only therapy had a 60% greater risk of ovarian cancer than participants who had never used hormone replacement therapy. A woman's risk increased by 7% with each year of use; it was nearly doubled after a decade of use and was tripled after two decades. Women who had undergone hysterectomy were also at significantly higher risk with long-term estrogen-only use. In contrast, women who had used only regimens containing both estrogen and progestin did not have a significantly elevated risk, although data were limited for this group.

The study was conducted in four phases: During the first (1979-1986), participants completed baseline and annual follow-up telephone interviews; during each subsequent phase (1987-1989, 1993-1995 and 1995-1998), they responded to a mailed questionnaire. Each survey asked women about their use of female hormones (injections, creams, patches or pills, depending on the study phase), menopausal status, gynecologic surgeries and risk factors.

Women enrolled in the study were former participants in a mammography screening project that took place in 29 study centers in the United States. Those who had had both ovaries removed, or had a history of breast or ovarian cancer, were excluded. A woman's follow-up began at the time of the initial telephone interview or menopause, whichever occurred later, and ended when she received a diagnosis of ovarian cancer, died, had both ovaries (or a second ovary) removed or completed the study's fourth phase.

New cases of ovarian cancer were ascertained from medical records, self-reports, state registry data and the National Death Index (with confirmation by review of death certificates). Rate ratios for ovarian cancer were generated by regression analysis for categorical variables and were adjusted for age, type of menopause (i.e., natural, surgical or unknown) and duration of oral contraceptive use. Score tests were used to assess the significance of trends in the risk of ovarian cancer according to duration of hormone use.

On average, the 44,241 women included in analysis were 56.6 years old at baseline. Ovarian cancer occurred in 329 women during a mean 13.4-year follow-up. The risk of ovarian cancer increased as women's number of live births and duration of oral contraceptive use decreased; it also was elevated among women who had not undergone hysterectomy. Risk was not associated with either age at menopause or body mass index.

Use of estrogen-only replacement therapy was more common among older than among younger women, among those who had had a hysterectomy than among those who had experienced natural menopause and among those who had gone through menopause before age 45 than among those who had done so later. Levels of use of estrogen-progestin compounds were relatively high among women who had experienced natural menopause, were older at menopause, had used oral contraceptives for more than two years and had a low body mass index.

Women who had ever used estrogen-only replacement therapy had a 60% increase in the risk of ovarian cancer, compared with women who had never used hormone replacement therapy (rate ratio, 1.6). The risk associated with estrogen-only use increased by 7% with each additional year of use, and the risk was sharply elevated after a decade of use. Women who had used estrogen-only replacement therapy for 10-19 years had a near doubling of risk (1.8), while those using it for 20 or more years had triple the risk of never-users (3.2).

Ovarian cancer risk was associated with use of hormone replacement therapy both for women who had had a hysterectomy and for those who had not. Among those who had had a hysterectomy before entering the study, the risk increased by a significant 8% with each year of estrogen-only use, and 20 or more years of use was associated with a tripling of risk (rate ratio, 3.4). Among women who had not had a hysterectomy at baseline, risk doubled after 4-9 years of use (2.1). In this group, longer-term estrogen-only use was uncommon.

Women who had used only combined hormone replacement therapy did not have a significantly higher risk of ovarian cancer than women who had never used replacement hormones; however, only 18 such women developed ovarian cancer, so the analysis was limited. The duration of use of this type of therapy (which averaged only 5.6 years among long-term users) was not significantly associated with the risk of ovarian cancer. Women who switched from estrogen-only to combined hormone therapy had a marginal increase in risk of ovarian cancer, which the researchers hypothesize reflected their prior use of estrogen-only therapy.

The study's results were not altered when analyses were further adjusted for family history of ovarian cancer (available for 71% of women) and family history of breast cancer.

Commenting on the findings, the investigators observe that many women who had used estrogen-only formulations were likely exposed to higher daily doses of hormone than are used today; the study could not determine the independent effects of dose and duration. Therefore, they conclude that it remains uncertain if long-term use of lower doses of estrogen is associated with an elevated risk of ovarian cancer. Likewise, because few cancers occurred in women using only combined hormone preparations and because long-term data for this use are lacking, "it is premature to conclude that [estrogen-progestin replacement therapy] has no association with ovarian cancer," the investigators caution. Confirmation of the study's findings, will require additional studies, they conclude, "with particular attention to duration, dose, and regimen."

This and other observational studies do not establish a causal link between use of estrogen replacement therapy and ovarian cancer risk, notes the author of an accompanying editorial.2 Nonetheless, he concludes that "the association between estrogen use and ovarian cancer should be worrisome enough for clinicians to consider carefully whether to suggest estrogen-only [hormone replacement therapy]."--S. London


1. Lacey JV et al., Menopausal hormone replacement therapy and risk of ovarian cancer, Journal of the American Medical Association, 2002, 288(3):334-341.

2. Noller KL, Estrogen replacement therapy and risk of ovarian cancer, editorial, Journal of the American Medical Association, 2002, 288(3):368-369.