摘要
Purpose: Few studies have investigated the clinical impact of whole-body positron emission tomography (PET) with <sup>18</sup> F-fluorodeoxyglucose (FDG) in endometrial cancer. We aimed to assess the value of integrating FDG-PET into the management of endometrial cancer in comparison with conventional imaging alone. Methods: All patients with histologically confirmed primary advanced (stage III/IV) or suspicious/documented recurrent endometrial cancer, with poor prognostic features (serum CA-125 >35 U/ml or unfavourable cell types), or surveillance after salvage therapy were eligible. Before FDG-PET scanning, each patient had received magnetic resonance imaging and/or computed tomography (MRI-CT). The receiver operating characteristic curve method with calculation of the area under the curve (AUC) was used to compare the diagnostic efficacy. Clinical impacts were determined on a scan basis. Results: Forty-nine eligible patients were accrued and 60 studies were performed (27 primary staging, 33 post-therapy surveillance or restaging on relapse). The clinical impact was positive in 29 (48.3%) of the 60 scans. Mean standardised uptake values (SUVs) of true-positive lesions were 13.2 (range 5.7-37.4) for central pelvic lesions and 11.1 (range 1.5-37.4) for metastases. The sensitivity of FDG-PET alone (P<0.0001) or FDG-PET plus MRI-CT (P<0.0001) was significantly higher than that of MRI-CT alone in overall lesion detection. FDG-PET plus MRI-CT was significantly superior to MRI-CT alone in overall lesion detection (AUC 0.949 vs 0.872; P=0.004), detection of pelvic nodal/soft tissue metastases (P=0.048) and detection of extrapelvic metastases (P=0.010), while FDG-PET alone was only marginally superior by AUC (P=0.063). Conclusion: Whole-body FDG-PET coupled with MRI-CT facilitated optimal management of endometrial cancer in well-selected cases. © Springer-Verlag 2005.