Background: Postmenopausal hormone therapy combining estrogen and progestin has been linked with an increased risk of breast cancer, but an association between breast cancer and estrogen-only hormone therapy is more controversial. A meta-analysis of 45 studies found no association between estrogen-only hormone therapy and breast cancer. Subsequent large studies reported an increased risk (relative risk [RR] = 1.30) for predominantly estradiol-based regimens but a reduced risk (RR = 0.77) for conjugated equine estrogens, even with more than six years of use. Fewer data are available on the association of cancer with trans-dermal estrogen preparations. Two studies with relatively short follow-up periods did not report increased risk of breast cancer. Lyytinen and colleagues used data from large population databases in Finland to examine the association between estrogen-only hormone therapy and breast cancer.

The Study: The authors used a national pharmacy database to identify women older than 50 years who bought at least one prescription of estrogen-only hormone therapy between 1994 and 2001. They excluded women who took conjugated equine estrogens and those who took estrogen for less than six months, leaving a study population of 110,984 women. Most (84,729) used systemic estradiol, but 7,941 used systemic estriol and 18,314 used vaginal estrogens. The cumulative exposure to estrogens was calculated for each woman. The women were monitored for development of breast cancer until the end of 2002 or death. Cancer diagnoses and deaths were abstracted from the comprehensive Finnish National Cancer Registry.

Results: During 648,022 patient-years of observation, there were 2,171 cases of breast cancer. For each group in the study, an expected number of cases was calculated based on the average incidence of breast cancer in all Finnish women. The ratio of observed number of cases to expected cases gave a standardized incidence ratio (SIR) for different patterns of estrogen use (see accompanying table).

Estradiol use for more than five years was associated with an increased risk of breast cancer (SIR = 1.44). The risk was also elevated in women recorded as having less than five years of exposure (SIR = 0.93 to 1.30), but many of these women may have been taking estrogens before the establishment of the pharmacy registry. The SIR increased with daily dose of estradiol and was significantly elevated in women who took more than 1.9 mg per day for five years or more. Use of patches or transdermal preparations of estradiol for five years or more also was associated with elevated SIRs. Oral estriol and vaginal estrogens were not associated with increased risk of breast cancer.

Conclusion: The authors estimate that use of oral or transdermal estradiol for five years or more is associated with two to three additional cases of breast cancer per 1,000 women during 10 years of follow-up.