Value Of ～(18)FDG-PET In The Diagnosis And Staging Of Colorectal Cancer

Objective: The present study was designed to assess the usefulness of positron emission tomography withfluorine-18-fluorodeoxyglucose (18F-FDG PET) for diagnosing and staging patients with colorectal cancer and to evaluate the value of standardized uptake value (SUV) and tumor to liver SUV ratio (TLR) in differentiating benign and malignant lesion.Methods: Twenty-five consecutive patients with primary or local recurrent colorectal cancer and fifteen with benign colorectal diseases or postoperative colorectal cancer were studied. Allpatients were performed by 18F-FDG PET and twenty-five by CT. The results of PET and CT were compared with a "gold standard" which consisted of surgical pathology, Colonoscopy or clinical and radiological follow-up. The sensitivity, specificity and accuracy of diagnosing primary or local recurrent cancer and metastatic tumor were calculated and compared between PET and CT. Forquantitative evaluation, a region of interest (ROI) was placed over the area of maximum activity within the lesions. A background ROI was then placed over the nontumor region of the liver. SUV and TLR was calculated and compared among primary or local recurrent cancer, metastatic tumor, benign lesions and normal intestines. Receivers operating characteristic (ROC) curves of SUV and TLR were generated for distinguishing between benign and malignant lesions.Results: When the visual method was used, the accuracy of 18FDG-PET for detecting primary and local recurrent colorectal cancer were 95.12%, whereas those of CT were 73.08% (PO.05).1 Q ____FDG-PET was found to be more specific and accurate than CT when compared with actual operative findings in diagnosing metastatic lesions in the abdomen as a whole (98.95% vs. 89.13% and 95% vs. 82.81%, PO.05). A significant correlation was found between SUVmax (or TLR) and SUVavg (or TLR) and between SUV and TLR of benign or malignant lesions. The SUV and TLR of primary or local recurrent cancer and metastatic lesions in the abdomen were significantly higher than that of benign lesions and normal intestines (p < 0.01). The SUVavg of primary or local recurrent tumor were slightly higher than that of metastatic lesionsin the abdomen (p < 0.05), while TLR and SUVmax were significantly higher than that of metastatic lesions (p < 0.01). The SUV of primary or local recurrent cancer had a significant correlation with their bulk. According to an ROC analysis, the areas under the ROC curves for SUVmax, SUVavg, TLRmax and TLRavg were 0.846, 0.872, 0.835 and 0.913, respectively. In the ROC curve, the best thresholds of SUVmax, SUVavg, TLRmax and TLRavg to distinguish benign from malignant lesions was 3.87, 2.38, 1.62 and 1.31, respectively.Conclusions: These results suggest that FDG-PET is useful not only for the evaluation of primary or local recurrent colorectal cancer but also for the diagnosis of metastatic lesions. The SUV and TLR presented a high value in distinguishing benign from malignant lesions.