Shengjie Xu,1,* Ridwan Abdifatah Mohamed,1,* Juan Xu,1,2 Yao Yao,3 Yun Gu,4 Wenwei Tang,3 Xuemei Jia1,2 

1Department of Gynecology, Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital, Nanjing, 210004, People’s Republic of China; 2Nanjing Medical Key Laboratory of Female Fertility Preservation and Restoration, Nanjing, 210004, People’s Republic of China; 3Department of Radiology, Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital, Nanjing, 210004, People’s Republic of China; 4Department of Pathology, Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital, Nanjing, 210004, People’s Republic of China

*These authors contributed equally to this work

Correspondence: Wenwei Tang, Department of Radiology, Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital, Nanjing, 210004, People’s Republic of China, Email tww3077@163.com Xuemei Jia, Department of Gynecology, Women’s Hospital of Nanjing Medical University, Nanjing Women and Children’s Healthcare Hospital, No. 123 Mochou Road, Nanjing, 210004, People’s Republic of China, Email xmjia@njmu.edu.cn

Purpose: The study aimed to evaluate the effectiveness of using specific indicators, particularly the apparent diffusion coefficient (ADC), alone or in combination to differentiate endometrial cancer (EC) from atypical endometrial hyperplasia (AEH) and to explore non-invasive biomarkers for the molecular classification of EC.
Methods: A retrospective analysis was conducted on 300 EC and 126 AEH cases who had undergone preoperative magnetic resonance imaging, complete blood count, coagulation profile testing, and tumor biomarkers assessment. Postoperative molecular classification was conducted on 76 EC samples. Diagnostic values were assessed using receiver operating characteristic (ROC) analysis and binary logistic regression with forward selection to determine the optimal indicator combinations. Furthermore, this study evaluated the variability of parameters across EC molecular subtypes.
Results: The ADC effectively balanced sensitivity and specificity in differentiating EC from AEH. An optimal diagnostic model including age, fibrinogen, and ADC achieved the area under the curve (AUC) of 0.9143, with 84.67% sensitivity and 88.89% specificity. ADC values were found to be lower in EC cases that exhibited a higher Ki-67 index or a higher histological grade. Notably, the NSMP subtype presented significantly higher ADC values compared to the other three molecular subtypes. The p53abn subtype exhibited the highest prevalence of abnormal HE4 levels and patients aged ≥ 65 (both 6/12, 50%) yet normal CA125 and CA19-9 levels.
Conclusion: This retrospective study demonstrated that ADC, especially when combined with age and fibrinogen, is a valuable biomarker for distinguishing EC from AEH. In addition to indicating the Ki-67 index and histological grade, ADC values also serve as a promising tool for identifying the NSMP subtype within EC. Future studies should focus on multi-center, prospective studies with larger sample sizes to validate and refine the diagnostic value of ADC in differentiating EC from AEH, as well as in the molecular classification of EC.

Keywords: endometrial cancer, atypical endometrial hyperplasia, apparent diffusion coefficient, histological grade, Ki-67 index, molecular subtype