Article Text
PDF
Abstract
Introduction/Background FDG-PET/CT is commonly used for preoperative staging of endometrial cancer (EC), but its accuracy in predicting EC lymph node metastasis requires clarification. This study aims to assess the agreement between preoperative FDG-PET/CT assessment of lymph node involvement and the definitive pathological analysis after EC surgery.
Methodology This retrospective multicentric international study will include approximately 100 EC patients who underwent preoperative FDG-PET/CT and either bilateral sentinel node sampling, pelvic, primary iliac, and lumbo-aortic lymph node sampling, or pelvic lymph node sampling with sentinel lymph node biopsy from January 2017 to April 2023. Carcinosarcoma subtypes will be excluded. A centralized review of FDG-PET/CT and pathology results will be conducted by two independent experts. Further analysis will occur at the end of inclusion, including subgroup analysis based on the EC histological subtype, grade, FIGO classification.
Results In the initial 43 cases from three centers (CHUV Lausanne and HFR Fribourg in Switzerland; Foch in France), a 21% discordance between FDG-PET/CT and postoperative anatomopathology of lymph nodes was observed, with a concordance of 79%. The sensitivity of FDG-PET/CT was 60%, specificity was 86%, positive predictive value was 37.5% and negative predictive value was 95%. The Area Under the Curve ROC was 73% with a standard error of 12%.
Conclusion Preliminary results show poor sensitivity of FDG-PET/CT in predicting lymph node involvement for EC. Definitive results will tell if the role of preoperative FDG-PET/CT should be rediscussed, as it actually strongly influence multidisciplinary tumor board decisions with a more extensive surgery and higher morbidity in case of false positive lymph nodes.
Disclosures We have no potential conflict of interest to report.
