RÉSUMÉ
PURPOSE: We aimed to assess the role of positron emission tomography using fluorodeoxyglucose (FDG-PET) in the diagnosis of recurrence or the assessment of therapeutic response in cervical and ovarian cancer patients through making a comparison between FDG-PET, abdominal computed tomography (CT) and serum tumor marker. Materials and METHODS: We included 103 cases (67 patients) performed FDG-PET and abdominal CT. There were 42 cervical cancers and 61 ovarian cancers. We retrospectively reviewed the interpretations of PET and CT images as well as the level of tumor marker. We calculated their sensitivity, specificity, positive predictive value and negative predictive value for these three modalities. And then we analyzed the differences between these three modalities. RESULTS: Tumor recurrences were diagnosed in 37 cases (11 cervical cancers and 26 ovarian cancers). For PET, CT and tumor marker, in cervical cancer group, sensitivity was 100% (11/11), 54.5% (6/11) and 81.1% (9/11), respectively. And specificity was 93.6% (29/31), 93.6% (29/31) and 100% (31/31). In ovarian cancer group, sensitivity was 96.2% (25/26), 84.6% (22/26) and 80.8% (21/26), and specificity was 94.3% (33/35), 94.3% (33/35), 94.3% (33/35). PET was highly sensitive to detect the intraperitoneal and extraperitoneal metastasis with the help of the CT images to localize the lesions. However, CT had limitations in differentiation of the recurrent tumor from benign fibrotic tissue, identification of viable tumors at the interface of tissues, and detecting extraperitoneal lesions. CONCLUSION: FDG-PET can be an essential modality to detect the recurrent or residual tumors in gynecologic cancer patients because of its great field of the application and high sensitivity.