Ann Surg 2013 Aug;258(2):289-95
Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY 10065, USA.
Objective: To prospectively compare the ability of flourodeoxyglucose-positron emission tomography (FDG-PET) and computed tomography (CT) to identify a pathological complete response (pCR) in patients with rectal cancer treated by chemoradiation.
Background: A major obstacle in pursuing nonoperative management in patients with rectal cancer after chemoradiation is the inability to identify a pCR preoperatively.
Methods: A total of 121 patients with rectal cancer were prospectively enrolled. FDG-PET scans and helical CT scans were obtained before and after neoadjuvant chemoradiation. Consensus readings of PET and CT scans were used to classify certainty of disease (5-point confidence rating scale). The ability of PET and CT scans to accurately distinguish a pCR (ypT0) from an incomplete response (ypT1-4) was estimated using the area under the receiver operating characteristic curve (AUC).
Results: Of the 121 patients, 26 (21%) had a pCR. PET and CT scans were equally inadequate at distinguishing a pCR from an incomplete response (AUC = 0.64 for both, P = 0.97). Among the 26 patients with a pCR, 14 (54%) and 5 (19%) were classified as complete responders on PET and CT scans, respectively. Among the 95 patients with an incomplete pathological response, 63 (66%) and 90 (95%) were classified as incomplete responders on PET and CT scans, respectively. None of the individual PET parameters, including visual response score, mean standard uptake value (SUVmean), maximum SUV (SUVmax), and total lesion glycolysis, accurately distinguished a pCR (AUCs = 0.57-0.73).
Conclusions: Neither PET nor CT scans have adequate predictive value to be clinically useful in distinguishing a pCR from an incomplete response and, therefore, should not be obtained for the purpose of attempting to predict a pCR after neoadjuvant chemoradiation for rectal cancer.