Histopathologic distinctions in the relationship of estrogens and endometrial cancer

R. I. Horwitz, A. R. Feinstein, R. A. Vidone, S. C. Sommers and S. J. Robboy

The slides of 233 patients included in a case-control study of estrogens and endometrial cancer were reviewed to determine how often endometrial cancer was misdiagnosed and whether patients with uterine cancer had other coexistent endometrial diseases. Reasonably close agreements were found among the original diagnoses and those of three additional reviewers (the total range of disagreements among all pathologists was from 2% to 16%). Proliferative and hyperplastic endometrium coexisted in many specimens from patients with endometrial cancer, and especially in those who had used estrogen replacement therapy. In contrast, estrogen therapy had seldom been used by patients whose cancers were not accompanied by these proliferative and hyperplastic lesions. In addition, these changes were found significantly more often in women with grade 1 cancers than grade 2 or 3 cancers. We conclude from these data that diagnostic misclassification is uncommon and that coexistent proliferative and hyperplastic lesions occur frequently, especially among women with grade 1 cancers. The data also suggest that the frequent finding of grade 1 cancer in estrogen users is due to bleeding that results from the stimulated coexistent benign proliferating endometrium.