A groundbreaking study has developed and validated a risk prediction model for endometrial hyperplasia (EH) and endometrial carcinoma (EC) aimed at premenopausal women, which could significantly inform clinical decisions.
Endometrial cancer, one of the most prevalent gynecological malignancies globally, has seen rising incidence rates due to factors like increasing obesity and declining fertility. The study, conducted at Chongqing Ninth People’s Hospital, sought to create quantitative standards and improve clinical outcomes for women with suspected EH and EC.
Data collected over seven years included 1541 premenopausal women who underwent hysteroscopic endometrial biopsy. The researchers created two groups for training (n = 1156) and validation (n = 385) of the predictive model.
Utilizing univariable and multivariable logistic regression analyses, the study identified independent risk factors such as body mass index (BMI), age at menarche, intrauterine device (IUD) use, diabetes, polycystic ovary syndrome (PCOS), endometrial thickness (ET), and the presence of uterine cavity fluid. These were incorporated to establish the nomogram model.
The nomogram exhibited strong discrimination and accuracy, with areas under the receiver operating characteristic curve (AUC) of 0.845 and 0.905 for the training and validation sets, respectively. The model was evaluated using decision curve analysis, which demonstrated its clinical utility.
Given the model's optimal cutoff for risk categorization, patients scoring below 76.411 are considered low risk and can pursue non-invasive treatments, whereas those exceeding this score are classified as high risk and should undergo invasive diagnostics.
Notably, the study emphasizes the necessity for individualized approaches to management, especially for young women concerned about the potential physical and psychological impacts of unnecessary invasive procedures. The findings indicate the model's practicality can substantially benefit physicians making decisions about patient management, potentially safeguarding reproductive health.
Enhancing patient care through predictive analytics, the research sets the stage for significant advancements in the treatment and management of EH and EC.
The study presents several limitations, including its retrospective nature and single-center design, indicating the need for future external validations across different medical settings. Plans for prospective studies are also underway to refine the model's accuracy and applicability.