Article Text
PDF
Abstract
Introduction/Background Nowadays, preoperative risk assessment for extrauterine disease in endometrial cancer (EC) patients is based on histotype and grade in preoperative biopsy, among others. Nevertheless, a poor correlation between preoperative and postoperative histological features has been observed. Therefore, we aimed to evaluate the performance of EC molecular classification to predict the presence of extrauterine disease during surgery when compared to other clinical and pathological information used in preoperative setting.
Methodology Retrospective single-center observational study developed at Vall d’Hebron University Hospital, Barcelona, Spain. Patients with endometrial adenocarcinoma treated with primary surgery between December-1994 and May-2022 were included for analysis. Molecular profiling was retrospectively performed in preoperative endometrial biopsies or hysterectomy specimens. Preoperative clinical, pathological and imaging information were reviewed, as well as postoperative FIGO staging, treatments and oncologic outcomes.
Results A total of 658 patients, 47(7.1%) with POLEmut tumors, 234(35.6%) with MMRd tumors, 95(14.4%) with p53abn tumors and 282(42.8%) with NSMP tumors. Extrauterine disease after primary surgery was diagnosed in 11.7% of EC patients, while a 10.7% presented positive LNs. P53abn tumors showed a significant increased proportion of extrauterine spread and nodal invasion when compared to the rest of molecular subgroups (34.1% and 30.1%, p<.001). In multivariate analysis only p53abn molecular subgroup (aOR=16.0, CI95%=1.5–165.1) and radiological suspicion of extrauterine disease (aOR= 24.2, CI95%=12.2–48.2) independently predicted the finding of extrauterine disease. In patients with preoperatively uterine-confined disease, predictive models suggested that myometrial invasion >50% in radiological assessment and p53abn molecular subtype where the best identifiers of occult disease diagnosed after the staging surgery.
Conclusion In EC patients, molecular profiling has the potential to anticipate the existence of extrauterine disease. When combined with imaging tests, it constitutes the most precise preoperative information for customizing surgical approaches.
Disclosures This study was funding via: ISCIII grant(PI20/00664,AC21_2/00030), Miguel Servet grant(CP22_00147), RETOS Colaboración(CPP2021–008440), FEDER, Fundación Científica AECC(GCTRA1804MATI and PERME212443COLA), Biomedical Research Center Network CIBERONC(CB16/12/00328) and Generalitat de Catalunya(2021SGR11757).
