Comparison of restaging accuracy of repeat FDG-PET/CT with pelvic MRI after preoperative chemoradiation in patients with rectal cancer

The aim of this study was to compare the restaging accuracy of repeat fluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) scan with pelvic magnetic resonance imaging (MRI) in patients with rectal cancer who have undergone preoperative chemoradiation. One hundred and eighty-one consecutive patients with locally advanced rectal cancer who underwent a total mesorectal excision after preoperative chemoradiation were prospectively enrolled. All the patients underwent FDG-PET/CT and pelvic MRI before chemoradiation and 5 weeks after the completion of chemoradiation. We evaluated the measurements of the FDG uptake (SUVmax) and the percentage of SUVmax difference (Response Index = RI) between the pre- and postchemoradiation FDG-PET/CT scans. The accuracy of repeat FDG-PET/CT and pelvic MRI for predicting pathologic CR were compared. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of pelvic MRI for predicting pathologic CR were 38.5, 58.1, 13.3, 84.9, and 55.2 %, respectively. In terms of FDG-PET/CT, pretreatment tumor size and pathologic stage were significantly correlated with the RI values. Using a RI value of 63.6 % as the cutoff threshold, it was possible to discriminate the CR from the non-CR with a sensitivity of 73.1 %, a specificity of 64.5 %, a PPV of 25.7 %, a NPV of 93.5 %, and an accuracy of 65.7 % (area under the curve = 0.723, 95 % confidence interval 0.619-0.828, P < 0.001). The accuracy of FDG-PET/CT restaging is superior to that of MRI staging for predicting pathologic CR in irradiated rectal cancer. An NPV of 93.5 % indicates that FDG-PET/CT can rule out the pathologic CR.