Introduction
Atypical endometrial hyperplasia (AEH) represents a critical juncture in gynecologic health, often serving as a precursor to endometrial cancer (EC). However, the diagnostic path is frequently complicated by the presence of "hidden" malignancies. Research indicates that nearly 30% of patients diagnosed with AEH preoperatively are found to have concurrent EC during their definitive hysterectomy. This diagnostic gap presents a major clinical hurdle: if cancer is not suspected beforehand, the initial surgery may be incomplete—lacking essential steps like lymph node staging. Such omissions can lead to suboptimal patient management, delayed adjuvant therapies, and potentially poorer long-term outcomes. Consequently, there is an urgent need for preoperative tools that can accurately predict the presence of concurrent EC in patients initially diagnosed with AEH.
Methodology
This retrospective study analyzed 455 patients diagnosed with preoperative AEH who underwent hysterectomy at a tertiary hospital between January 2020 and December 2022. The researchers aimed to determine if preoperative levels of human epididymis protein 4 (HE4) could predict the presence of concurrent EC.
Using a multivariate logistic regression model, the team identified independent predictive factors. These factors were then used to construct a nomogram—a graphical tool designed to estimate the probability of concurrent EC for individual patients. The model's reliability was assessed through internal validation using a bootstrap resampling method.
Key Findings
The study confirmed that concurrent EC is a prevalent issue, occurring in 23.4% of the cohort.
The Role of HE4
Preoperative HE4 levels were significantly higher in patients with concurrent EC compared to those with confirmed AEH (median 50.5 pmol/L vs. 43.7 pmol/L, p < 0.001). Receiver operating characteristic (ROC) analysis demonstrated that HE4 has strong predictive potential:
- Concurrent EC: AUC = 0.696 (95% CI = 0.633–0.760)
- Intermediate-to-High-Risk EC: AUC = 0.713 (95% CI = 0.563–0.863)
Independent Risk Factors
Multivariate analysis identified three primary independent predictors for concurrent malignancy:
- Postmenopausal Status: (OR = 5.25)
- HE4 Level: (OR = 3.84)
- Body Mass Index (BMI): (OR = 2.09)
Nomogram Performance
By combining these three factors, the researchers developed a nomogram that outperformed any single predictor. The bootstrap-corrected concordance index (C-index) for the nomogram was 0.725 (95% CI = 0.665–0.784). The model showed high consistency between predicted probabilities and observed clinical outcomes, suggesting significant potential for clinical utility.
Clinical Implications
The development of this HE4-based nomogram represents a significant step toward personalized surgical planning in gynecologic oncology. By moving beyond a "one-size-fits-all" approach to AEH, clinicians can now leverage a combination of biochemical markers and patient demographics to assess risk. The synergy between HE4, BMI, and menopausal status captures a broader picture of the patient's physiological state than any single metric.
In practice, this tool allows for a more aggressive, cancer-appropriate surgical approach to be planned from the outset for high-risk individuals. This proactive strategy minimizes the psychological and physical burden of repeat surgeries on patients and ensures that those with concurrent EC receive the full standard of care without delay. While the study's retrospective nature and single-center data suggest the need for future external validation, the current results provide a strong foundation for improving the preoperative stratification of AEH patients.
Conclusion
Preoperative HE4 is a valuable biomarker for identifying concurrent endometrial cancer in patients initially diagnosed with AEH. The developed nomogram, incorporating HE4, postmenopausal status, and BMI, offers a practical and effective tool for improving surgical strategy and patient management in gynecologic oncology.
Summary updated