A risk-stratified model for predicting endometrial atypical hyperplasia and cancer to guide biopsy decisions in asymptomatic postmenopausal women.

Background: Endometrial atypical hyperplasia (EAH) and endometrial cancer (EC) are increasingly detected in asymptomatic postmenopausal women. This often leads to delayed treatment. Risk stratification remains challenging, and single-factor models may not accurately identify high-risk individuals. This study aimed to develop and validate a multivariable prediction model for identifying EAH or EC in asymptomatic postmenopausal women. Methods: This retrospective cohort study included asymptomatic postmenopausal women with endometrial pathology records from the Third Affiliated Hospital of Sun Yat-sen University, China (2021-2024). Candidate risk factors included demographics, clinical characteristics, and hematological parameters. The primary outcome was a composite of histologically-confirmed EAH or EC. Multivariable Poisson regression with robust variance was then employed to identify independent risk factors for this composite outcome. Risk - stratified models were developed by calculating predicted probabilities for key combinations of risk factors. Results: Among 928 patients [median age: 59 years, IQR (interquartile range): 55-65; median BMI: 23.4 kg/m2], the overall prevalence of EAH and EC was 2.59% (24/928). Key independent risk factors included endometrial thickness (EMT) > 8 mm [vs. ≤ 4 mm: adjusted Relative Risks (aRR) = 11.34, 95% confidence interval (95% CI): 4.35-39.56; p < 0.001], diabetes (aRR = 2.54, 95% CI: 1.12-5.78; p = 0.026), and platelet count per 109/L increase (aRR = 1.01, 95% CI: 1.01-1.02; p < 0.001). EMT > 8 mm was the strongest predictor, with the highest aRR. In the stratified analysis, the combination of EMT > 8 mm and diabetes was associated with the highest observed prevalence (33.33%). The risk-stratified model demonstrated clinical utility: using a ≥ 5% risk threshold, biopsy would be recommended for 17% of patients (sensitivity 70.8%, specificity 84.4%); at a ≥ 10% threshold, biopsy would be recommended for 4.3% of patients (requiring 3.64 biopsies per true positive case). Discussion: The risk of EAH and EC among asymptomatic postmenopausal women varies significantly based on clinical factors. This risk-stratified modeling approach delivers individualized risk estimates to inform endometrial biopsy decisions, facilitating personalized patient management.