F-18 FDG PET/CT in the assessment of patients with unexplained CEA rise after surgical curative resection for colorectal cancer.

Abstract PURPOSE: We evaluated the role of quantitative assessment by maximum standardized uptake value (SUVmax) on F-18 fluorodeoxyglucose [F-18]FDG positron emission tomography/computed tomography (PET/CT) in stratifying colorectal cancer (CRC) patients with unexplained carcinoembryonic antigen (CEA) rise after surgical curative resection. MATERIAL AND METHODS: Forty asymptomatic patients (mean age, 64 ± 12 years) with previous CRC and current serum CEA levels >5 ng/ml underwent [F-18] FDG PET/CT 13 ± 3 months after complete surgical resection. The SUVmax was registered on anastomosis and peri-anastomotic tissue lesions, if present. The patients were followed for 24 ± 9 months thereafter. Re-intervention, evidence of newly discovered distant metastases, and death were recognized as main events and constituted surrogate end points. The receiver-operator-curve (ROC) analysis was performed to estimate the optimal SUVmax cut-off to predict patients at high risk of main events. PET/CT results were then related to disease outcome (overall survival; OS). RESULTS: The mean SUVmax at the anastomotic site was 6.2 ± 3 (range 2.6-15). At multivariate logistic regression analysis, the anastomotic SUVmax remained as the only significant contributor to the prediction of the events (p = 0.004; OR 1.97). The ROC analysis recognized that the optimal threshold of SUVmax to differentiate patients was 5.7. A worse OS was observed in patients presenting with a SUVmax greater than 5.7 as compared to those having lesser (median survival: 16 vs. 31 months; p = 0.002). CONCLUSIONS: The quantitative assessment by SUVmax on [F-18]FDG PET/CT may be helpful in patients presenting with unexplained CEA rise after curative resection of CRC, by identifying those at risk of main events.