Background: Fluorine-18-2-fluoro-2-deoxy-D-glucose positron emission tomography (FDGPET) is frequently used in the diagnosis of tumors, including colorectal cancer (CRC). This study aimed to determine whether PET could be used to detect CRC and colorectal adenomas, with endoscopic and pathologic confirmation. Methods: The PET results were verified by colonoscopy. Patients who underwent total colonoscopy (TCS) and PET between October 2002 and September 2012 at a tertiary cancer center were evaluated. TCS and PET were performed independently within 1 year. Cases of diminutive polyps (1-5 mm) were not included. Advanced CRC was defined as tumor extending into the muscularis propria or deeper, corresponding to T2-T4 stages according to TNM classification. Early-stage CRC was defined as tumors invading within submucosal layer, corresponding to Tis and T1 stages. Excluding advanced CRC, lesions were classified according to morphology as flat, depressed, or protruded. PET and PET-CT images were reviewed by at least 2 board-certified radiologists. Focal FDG uptake was determined in each portion of the colon (cecum and ascending, transverse, descending, and sigmoid colon) and rectum. FDG uptake in a matched segment containing a lesion was defined as a true positive result. Cases with a previous history of colorectal resection, direct invasion or peritoneum dissemination of other cancers, hyperglycemia (>150 mg/dl), incomplete TCS, hereditary colon cancer, preceding therapy, and anal canal tumors were excluded. To clarify the detectability of adenoma and CRC, nonepithelial tumors were excluded. Results: We identified 2483 consecutive eligible patients. A total of 694 lesions, 6 mm or larger, were detected by TCS: 205 advanced CRCs, 132 earlystage CRCs, and 357 adenomas. Of the 694 lesions that were 6 mm or larger, 374 (53.9%) were PET positive. Of the 531 cases with positive FDG uptake, 374 (70.4%) had lesions detected in the matched segments. False positive, which was defined as no or trivial findings despite of significant FDG uptake, was 157 (29.6%). Sensitivities of PET for advanced CRC, early-stage CRC, and adenoma were 98%, 69%, and 23%, respectively. Table1 summarizes the sensitivities of PET stratified by tumor pathology and size. The increase in sensitivity correlated with lesion size and tumor grade. The sensitivity was nearly 50% for small (6-9 mm) early-stage CRCs. PET was able to detect approximately 40% of large adenomas (1019 mm in size); however, its sensitivity for small adenomas (6-9 mm) decreased to 10%. Regarding morphology, PET was significantly less sensitive for flat lesions than for depressed or protruded lesions (p < 0.001). Conclusions: The sensitivity of PET in diagnosis of CRC stratified by lesion size, pathology, and morphology was evaluated. PET had an acceptable sensitivity for early-stage CRCs and large adenomas. Table1. PET sensitivity stratified by lesion size and pathology
Read full abstract