EP138/#520  Prediction of final pathology depending on preoperative myometrial invasion and grade assessment in low risk endometrial cancer patients: a Korean gynecologic oncology group ancillary study

• Published in: International journal of gynecological cancer Vol. 33; no. Suppl 4; p. A134
• Main Authors: Lee, Bang-Hyun, Kang, Sokbom, Kim, Jong-Hyeok, Kim, Byoung Gie, Kim, Jae-Weon, Kim, Moon-Hong, Chen, Xiaojun, No, Jae-Hong, Lee, Jong-Min, Kim, Jae-Hoon, Watari, Hidemichi, Kim, Seok Mo, Kim, Sunghoon, Seong, Seok Ju, Jeong, Dae Hoon, Kim, Yun Hwan
• Format: Journal Article
• Language: English
• Published: Oxford BMJ Publishing Group Ltd 07.11.2023; BMJ Publishing Group LTD
• Subjects: Biopsy; Endometrial cancer; ePoster Viewing; Fertility; Oncology; Pathology
• ISSN: 1048-891X; 1525-1438
• DOI: 10.1136/ijgc-2023-IGCS.234

IntroductionFertility-sparing treatment might be considered option for reproductive women with low risk endometrial cancer (EC). However, in low risk EC patients, concordance rates between preoperative assessment and postoperative pathology are not high enough. We aimed to predict postoperative pathology depending on preoperative myometrial invasion (MI) and grade in low risk EC patients to help extend current criteria for fertility-sparing treatment.MethodsIn Korean Gynecologic Oncology Group (KGOG) 2015, a prospective, multicenter study, 529 EC patients underwent preoperative assessment using MRI and endometrial biopsy followed by surgical staging. This ancillary study included patients who had no MI or MI <1/2 on preoperative MRI and endometrioid adenocarcinoma and grade 1 or 2 on endometrial biopsy. Among eligible patients, Groups 1 - 4 were defined with no MI and grade 1, no MI and grade 2, MI <1/2 and grade 1, and MI <1/2 and grade 2, respectively. New prediction model using machine learning was developed.ResultsAmong 251 eligible patients, Groups 1 - 4 included 106 (42.2%) patients, 41 (16.3%), 74 (29.5%), and 30 (12.0%), respectively. Compared with conventional analysis, new prediction model showed somewhat better prediction values. In new prediction model, NPV, sensitivity, and AUC of preoperative each group to predict postoperative each group were 88.9%, 77.6%, and 0.714 for Group 1, 97.1%, 64.3%, and 0.676 for Group 2, 77.5%, 76.5%, and 0.641 for Group 3, and 92.4%, 64.9%, and 0.691% for Group 4.Conclusion/ImplicationsIn low risk EC patients, prediction of postoperative pathology was ineffective enough. New prediction model might provide better prediction.