Prognostic role of FDG PET-derived parameters in preoperative staging of endometrial cancer

Purpose: To investigate the preoperative prognostic role of 18F-FDG PET/CT in patients with endometrial carcinoma (EC). Methods: 18F-FDG PET/CT was performed in 57 patients for EC preoperative staging. Maximum and mean standardized uptake values (SUVmax, mean), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of primary tumors, at different thresholds of 40%, 50%, 60% (40–50–60), were evaluated and compared with anatomopathological features. The diagnostic performance of PET-parameters (categorized by ROC analysis) in discriminating low-intermediate and high-risk disease and the prognostic role on survival (overall survival –OS; disease free survival – DFS) was evaluated. Results: The categorized TLG40-50-60 were the only parameters related to FIGO stage I versus II-III-IV (p = 0.0035 for all). The cut-off values for risk stratification were 83.69, 61.81 and 41.32, respectively (sensitivity: 60.00%; specificity; 71.43% for all parameters). Pathological stage 1 (pT1) of the primary tumor was predicted by MTV60 and TLG40-50 (p = 0.0328, 0.0240, 0.0147, respectively). The optimal thresholds were 7.795, 99.55 and 77.58, respectively (sensitivity: 38.46%, 53.85% and 53.85%, respectively; specificity: 88.64%, 79.55% and 81.82%, respectively). SUVmax and SUVmean40-50-60 were the only parameters discriminating endometrioid from non-endometrioid subtype. The corresponding sensitivity was 64.86% and 62.16% for SUVmax and SUVmean 50-60 and 62.16% for SUVmean40; specificity was 70.00% for all parameters. The mean (SD) OS was 79.77% (3.34%) and the mean DFS was 77.89% (3.73%). The tumor type was the only variable significantly associated with OS (p = 0.0486). TLG50 > 77.58 cm3 was the only variable associated with a higher risk of relapse (p = 0.0472). Conclusion: TLG40–50–60 and MTV60 of primary EC have prognostic value in discriminating FIGO and pathological staging. These results suggest a possible role of these parameters in predicting EC aggressiveness, thus improving the preoperative characterization of endometrial cancer.