Background: Although effective at detecting locally recurrent colorectal cancer, the accuracy of fluorine-18 fluorodeoxyglucose positron emission tomography (FDG-PET) for detecting rectal cancer recurrence in an irradiated pelvis has not been systematically studied.
Study design: Records of surgically resected rectal cancer patients who underwent FDG-PET imaging at least 6 months after external beam radiation therapy (EBRT) were reviewed. Cases (n = 19) were defined as scans from patients in whom a pelvic recurrence was confirmed (histologically, n = 14, radiologic followup, n = 5). Controls (n = 41), defined as scans from patients without clinical or radiologic evidence of pelvic recurrence, were compared with cases for the time interval between completion of EBRT and FDG-PET imaging (RT/PET interval, mean 25.1 months versus 27.5 months, respectively), as well as EBRT dose (mean 5,084 cGy versus 5,062 cGy, respectively). All 60 FDG-PET scans were iteratively reconstructed and reinterpreted by a single nuclear medicine physician blinded to original FDG-PET interpretation and disease status. Certainty of disease was scored on a five-point scale (1 to 5), with scores greater than or equal to 4 considered positive.
Results: FDG-PET correctly identified 16 of 19 recurrences, for a sensitivity of 84% and specificity of 88%. Overall accuracy was 87%. Positive predictive value was 76% and negative predictive value was 92%. Positive predictive value and accuracy improved in scans performed more than 12 months after EBRT.
Conclusions: Our preliminary data suggest that FDG-PET is an accurate modality for detecting pelvic recurrence of rectal cancer after full-dose EBRT. Its reliability appears to improve with time, perhaps because of resolution of early postradiation inflammation.