Endometrial cancer and estrogen use. Report of a large case-control study

N Engl J Med. 1979 Jan 4;300(1):9-13. doi: 10.1056/NEJM197901043000103.


Our case-control study of the relation between estrogen use and endometrial cancer involved 451 cases and 888 controls. The overall risk of endometrial carcinoma was sixfold for estrogen users as compared with nonusers; long-term users (greater than five years) had a 15-fold risk. Excess risk was present for both diethylstilberstrol and conjugated estrogens. The risk associated with cyclic use was as great as that for continuous use. Increased risk was associated with estrogen use for all histologic grades of the tumor. The risk of advanced-stage carcinoma was fourfold for estrogen users, but rhe confidence interval was wide, and this question requires further study. Finally, this investigation contradicts the speculation that the association between this cancer and estrogen use can be explained by swifter diagnosis for estrogen users, misclassification of estrogen-related hyperplasia or treatment of early symptoms of the tumor with estrogen.

PIP: Comparing cases of endometrial cancer (EC) (451) to matched hospital controls (888), the estimated risk of EC associated with estrogen (E) use in this large-scale study was 6, with a 95% confidence interval of 3.7-9.7; this figure jumped to 15 for long-term E users ( 5 years). The case-control study found that excess risk was present for both diethylstilbestrol and conjugated Es. Cancer risk was associated about equally with cyclic and continuous dosage regimens, and all histologic grades of tumors were associated with increased risk with E use. For State I tumors, all 3 histological grades ("well-differentiated"-"poorly differentiated") were found with E use. The relative risk of advanced-stage carcinoma was 4 for E users, but the confidence interval was wide, and this question remains unclear. Questions of methodology are addressed which apply to all studies of E use and EC; these include suspicions that women under treatment with E receive swifter diagnoses of carcinoma, the misclassification of E-related hyperplasia, and the treatment of early symptoms of the tumor with E.

Publication types

  • Clinical Trial
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Clinical Trials as Topic
  • Diethylstilbestrol / adverse effects
  • Drug Administration Schedule
  • Endometrial Hyperplasia / pathology
  • Epidemiologic Methods
  • Estrogens / adverse effects*
  • Estrogens, Conjugated (USP) / adverse effects
  • Female
  • Humans
  • Maryland
  • Menopause
  • Middle Aged
  • Neoplasm Staging
  • Uterine Neoplasms / chemically induced*
  • Uterine Neoplasms / epidemiology
  • Uterine Neoplasms / pathology


  • Estrogens
  • Estrogens, Conjugated (USP)
  • Diethylstilbestrol