Endometrial cancer in relation to patterns of menopausal estrogen use.

Female residents of King County, Washington, in whom endometrial cancer developed between January 1975 and April 1976 were interviewed concerning prior use of menopausal estrogens. Their responses were compared with those of a random sample of women from the same population. Among current estrogen users, endometrial cancer risk was strongly related to duration of use; although only a minimal elevation of risk was present during the first two years, there was a rapid rise to a 20-fold excess after about ten to 15 years. Cessation of estrogen use led to a decline in incidence of endometrial cancer within several years, but the risk remained higher than in nonusers through the first decade after administration of the drug was stopped. Risk was elevated whether or not the regimen was cyclic and whether conjugated or other types of estrogens had been used. Dosages of less than 0.625 mg/day of conjugated estrogens produced a smaller increase in risk than did other dosages.