RESUMO
OBJECTIVE: This study aims to ascertain if semiquantitative measurements derived from F-fluorodeoxyglucose positron emission tomography/computed tomography can be used as prognostic markers in patients with newly diagnosed endometrial cancer. MATERIALS AND METHODS: Patients with endometrial cancer and a preoperative F-fluorodeoxyglucose positron emission tomography/computed tomography before curatively intended treatment were included. The scans were evaluated using standard uptake values [maximum standard uptake value (SUVmax) and partial volume corrected (c) mean standardized uptake value (SUVmean)] and whole-body total lesion glycolysis (cTLG). All measurements were analyzed as prognostic factors in relation to overall survival (OS). Receiver operating characteristic curves were performed on all 3 positron emission tomography measurements to find the optimal cut-off for predicting OS. Multivariate Cox proportional regression models were used for prognostic evaluation. RESULTS: Eighty-three patients (median age, 69.9 y; range, 26.8-91.1) with primarily high-risk endometrial cancer or suspected high The International Federation of Gynecology and Obstetrics stage were included. Mean follow-up time was 3.48 years (range, 0.31-6.87), and 24 patients died during follow-up. In multivariate analyses with adjustment for other known prognostic factors, a SUVmax of greater than or equal to 14.3 g/mL and cSUVmean of greater than or equal to 12.7 g/mL of the primary tumor yielded a hazard ratio for OS of 3.18 (1.19-8.49) and 1.93 (0.80-4.68), respectively. Whole-body cTLG of greater than or equal to 176.1 g yielded a hazard ratio of 5.70 (1.94-16.78) for OS in a multivariate analysis. CONCLUSIONS: Preoperative SUVmax and cTLG showed potential as independent prognostic markers of OS in patients with primarily high-risk endometrial cancer. Thus, SUVmax and cTLG might help identify patients who could benefit from a more aggressive treatment strategy or closer surveillance.