# Human Epididymis Protein 4 Predicted Concurrent Intermediate-high-risk Endometrial Cancer and Eligibility of Fertility-sparing Treatment for Patients Diagnosed with Endometrial Atypical Hyperplasia Before Surgery

## TL;DR

This study shows that a blood test for HE4 can help identify women with endometrial atypical hyperplasia who are at higher risk for cancer and less likely to benefit from fertility-preserving treatments.

## Contribution

The study introduces HE4 as a novel biomarker for predicting cancer risk and fertility treatment eligibility in preoperative-EAH patients.

## Key findings

- 26.2% of preoperative-EAH patients had concurrent endometrial cancer, with 6.0% classified as intermediate-high-risk.
- HE4 and postmenopausal status were independent predictors of intermediate-high-risk EC, leading to a nomogram with strong predictive accuracy (C-index 0.819).
- HE4 predicted non-candidates for fertility-sparing therapy in young preoperative-EAH women with high statistical significance.

## Abstract

Objective: To investigate whether serum human epididymis protein 4 (HE4) could identify concurrent intermediate-high-risk endometrial cancer (EC) in patients diagnosed with endometrial atypical hyperplasia before definitive surgery (preoperative-EAH).

Methods: This retrospective study analyzed preoperative-EAH patients who underwent hysterectomy at a tertiary hospital between January 2016 and December 2022.

Results: Among 715 preoperative-EAH patients, 26.2% (187/715) were diagnosed with concurrent EC postoperatively, with 6.0% (43/715) identified as having concurrent intermediate-high-risk EC. Serum HE4 and postmenopausal status were revealed as independent predictors of concurrent EC. Receiver operator characteristic analyses determined optimal HE4 cut-off values of 43.50 pmol/L for predicting concurrent EC, 53.15 pmol/L for intermediate-high-risk EC, and 43.80 pmol/L for identifying non-candidates for fertility-sparing treatment. Multivariate analyses confirmed HE4 and postmenopausal status as key independent predictors of intermediate-high-risk EC, leading to the development of a nomogram model. It demonstrated a bootstrap-corrected C-index of 0.819 (95% confidence interval [CI] = 0.74-0.90). The calibration and decision curves highlighted its consistency and clinical utility. According to the nomogram, 41.4% (24/58) of high-score patients had concurrent intermediate-high-risk EC, compared with only 2.9% (19/657) in the low-score group (P < 0.001). HE4 also significantly predicted the non-candidates for fertility-preserving therapy in young preoperative-EAH women (odds ratio [OR] = 5.21, 95% CI = 2.10-12.89, P < 0.001).

Conclusion: Serum HE4 was a promising predictor of concurrent intermediate-high-risk EC and suitability for fertility-sparing treatment for preoperative-EAH patients. Incorporating HE4 and menopausal status into the nomogram model significantly enhanced the risk stratification for intermediate-high-risk EC and might assist clinical decision-making.