Integrated histological parameters define prognostically relevant groups in atypical endometrial hyperplasia/endometrioid intraepithelial neoplasia.

Objective: To assess the risk of endometrial carcinoma following a diagnosis of atypical hyperplasia/endometrioid intraepithelial neoplasia by endometrial biopsy, stratified based on integrated histological parameters.
Methods: All women with atypical hyperplasia/endometrioid intraepithelial neoplasia undergoing hysterectomy within 1 year of diagnosis without progestin treatment were included. Patients were subdivided into three study groups, based on two criteria: (a) grade of nuclear atypia and (b) foci (<2 mm) of confluent glands with no intervening stroma: low-grade, high-grade, and confluent glands. The rate of endometrial carcinoma on the subsequent hysterectomy was assessed in each study group, and differences between study groups were assessed using Fisher's exact test, with a significant p value <0.05. Reproducibility was assessed by using Cohen's κ.
Results: Ninety-six patients were included. Overall, 36 of 96 patients (37.5%) had endometrial carcinoma on the subsequent hysterectomy. The number of endometrial carcinomas was 4 of 42 (9.5%) in the low-grade group, 14 of 28 (50.0%) in the high-grade group, and 18 of 26 (69.2%) in the confluent glands group. The rate of endometrial carcinoma was significantly higher in the high-grade group than in the low-grade group (p<0.001), whereas it did not significantly differ between the high-grade group and the confluent glands group (p=0.176). The reproducibility among pathologists was moderate for low-grade versus high-grade (κ=0.58) and substantial for confluent glands versus low-grade (κ=0.63) and high-grade (κ=0.63).
Conclusion: Atypical hyperplasia/endometrioid intraepithelial neoplasia can be stratified into prognostically relevant groups based on integrated histological parameters, with a possible major impact on patient management.
Competing Interests: Competing interests: None declared.
(© IGCS and ESGO 2024. No commercial re-use. See rights and permissions. Published by BMJ.)