Estrogen and endometrial cancer: Cases and two control groups from North Carolina

Abstract Endometrial cancer risk associated with exposure to exogenous estrogen was evaluated for 256 cases, 224 gynecology controls, and 321 community controls. The pathologic diagnosis on all cases and a sample of gynecology controls was verified independently by two outside pathologists. Data on estrogen use were obtained from interviews of patients or their relatives, physicians' office records, and hospital records. Consistent patterns of relative risks appeared with both control groups. Black women revealed no differences in estrogen use between cases and either set of controls. White women exhibited risks of 3.6 and 4.1 after 3½ years or more of estrogen use. No increased risk was observed for shorter durations of use, irrespective of dose, type of estrogen, mode of administration, stage or grade of cancer. Obese women, who presumably produce increased amounts of endogenous estrogen, did not have their risk increased further by exogenous estrogen. Nonobese, nonhypertensive women had the highest risks of 5 to 8 after 3½ years or more of estrogen use. A minimum latency period of 3 to 6 years subsequent to the first administration of estrogen was demonstrated, and an estrogen-free interval of 2 years was sufficient to reduce elevated risks to nonuser levels. These short latency and recency intervals are consistent with carcinogenic promoter rather than an initiator action. With estrogen use of long duration, risks of cancer were high for Stage IA and grade 1 lesions, and either small or not significant for more advanced disease. Increased medical surveillance that resulted in earlier diagnosis, or less aggressive biologic lesions, may account for these findings.