1891 women given conjugated estrogens for the menopause were followed for 12 years (mean) for incidence of breast cancer. Overall, 49 cases were observed; 39.1 were expected on the basis of rates in the general population (relative risk = 1.3, P = 0.06). The relative risk increased with follow-up duration, progressing to 2.0 after 15 years (13/6.6, P = 0.01). The excess risk after 10 years was not due simply to prolonged estrogen use, since there was no clear dose-response relation to accumulated years of use. However, higher risk accrued to women using higher-dose tablets and those taking the medication on an other than daily basis. In addition, after 10 years of follow-up observation, two factors related to low risk of breast cancer, multiparity and oophorectomy, were no longer so related. Finally, estrogen use was related to an especially high risk of breast cancer among women in whom benign disease developed after they had started the drug.
PIP: The average age of 1891 women at the start of estrogen therapy was 49 years; long-term replacement was the usual goal. At the end of the study (mean, 12 years) 1573 women were alive, 132 had died, and 186 (9.8%) were lost to the study. The study involved 22,717 person-years. Breast cancer developed in 49 women during this period vs. 39.1 expected on the basis of rates among similar women in the general population, a relative risk of 1.3. The relative risk increased with duration of follow-up, being .9 for the first 5 years after starting estrogen, 1.2 from 5 to 9 years, 1.3 from 10 to 14 years, and 2.9 after 15 years. This trend was considered statistically significant (p less than .02). The excess risk became manifest after about 12 years. Reduced risk with increased parity was noted during the first 10 years only. Most of these patients had undergone hysterectomy and bilateral oophorectomy. The increased breast cancer risk was unrelated to ovarian status. Those with a history of benign breast tumor showed a higher relative risk than others, particularly if the benign disease had developed after estrogen was given. Although a dose-response relation was not observed, the risk was highest among those who took the medication in other than a daily regimen. Findings suggest that menopausal estrogen therapy raises the risk of breast cancer. Data indicate that evaluation of these agents will necessitate the incorporation of several factors lacking in previous studies, e.g., latent period risk indicators, type of therapeutic regimen, and total accumulated dosage.