PURPOSE: We evaluated the ability of pretreatment (18)F-FDG uptake by regional lymph nodes to predict the survival of patients with resectable colorectal cancer. METHODS: The records of 78 patients with AJCC stage III colorectal cancer (pathologically confirmed node-positive disease without evidence of distant metastasis) treated with surgery and adjuvant chemotherapy were retrospectively reviewed. The maximum standardized uptake values of the primary tumor (SUVp) and regional lymph nodes (SUVn) were measured by pretreatment (18)F-FDG PET/CT. The ROC curve analyses and the Cox proportional hazard model were used to analyze whether SUVp, SUVn, and clinicopathologic parameters could predict disease-free survival. RESULTS: Although there were no significant differences between the median SUVp in the event group and that in the non-event group, the median SUVn was significantly higher in the event group (1.7) than in the non-event group (0.8, p = 0.023). Based on the ROC curve analysis, SUVn predicted the event for disease-free survival (AUC = 0.668, p = 0.02) with the optimal criterion, sensitivity, specificity, and accuracy of > 1.2, 71%, 63%, and 65%, respectively. However, SUVp did not predict disease-free survival (AUC = 0.570, p = 0.349). Univariate analysis revealed that SUVn (p = 0.011) and venous invasion (p = 0.016) were associated with disease-free survival, but pathologic N stage was not (p = 0.09). By multivariate analysis, only SUVn > 1.2 independently shortened the disease-free survival (relative risk, 2.97; 95% CI, 1.14-7.74, p = 0.026). CONCLUSION: SUVn before surgery may be a useful prognostic marker in patients with AJCC stage III colorectal cancer.
PURPOSE: We evaluated the ability of pretreatment (18)F-FDG uptake by regional lymph nodes to predict the survival of patients with resectable colorectal cancer. METHODS: The records of 78 patients with AJCC stage III colorectal cancer (pathologically confirmed node-positive disease without evidence of distant metastasis) treated with surgery and adjuvant chemotherapy were retrospectively reviewed. The maximum standardized uptake values of the primary tumor (SUVp) and regional lymph nodes (SUVn) were measured by pretreatment (18)F-FDG PET/CT. The ROC curve analyses and the Cox proportional hazard model were used to analyze whether SUVp, SUVn, and clinicopathologic parameters could predict disease-free survival. RESULTS: Although there were no significant differences between the median SUVp in the event group and that in the non-event group, the median SUVn was significantly higher in the event group (1.7) than in the non-event group (0.8, p = 0.023). Based on the ROC curve analysis, SUVn predicted the event for disease-free survival (AUC = 0.668, p = 0.02) with the optimal criterion, sensitivity, specificity, and accuracy of > 1.2, 71%, 63%, and 65%, respectively. However, SUVp did not predict disease-free survival (AUC = 0.570, p = 0.349). Univariate analysis revealed that SUVn (p = 0.011) and venous invasion (p = 0.016) were associated with disease-free survival, but pathologic N stage was not (p = 0.09). By multivariate analysis, only SUVn > 1.2 independently shortened the disease-free survival (relative risk, 2.97; 95% CI, 1.14-7.74, p = 0.026). CONCLUSION: SUVn before surgery may be a useful prognostic marker in patients with AJCC stage III colorectal cancer.
Authors: U I Attenberger; L R Pilz; J N Morelli; D Hausmann; F Doyon; R Hofheinz; P Kienle; S Post; H J Michaely; S O Schoenberg; D J Dinter Journal: Eur J Radiol Date: 2014-03-22 Impact factor: 3.528
Authors: K K Chan; B Dassanayake; R Deen; R E Wickramarachchi; S K Kumarage; S Samita; K I Deen Journal: World J Surg Oncol Date: 2010-09-15 Impact factor: 2.754
Authors: Jang Eun Lee; Sang Woo Kim; Jin Su Kim; Kyu Yong Choi; Won Kyung Kang; Seong Taek Oh; Ie Ryung Yoo; Sung Hoon Kim Journal: World J Gastroenterol Date: 2012-09-28 Impact factor: 5.742