대한대장항문학회

Purpose: The prognosis of advanced colorectal cancer patients may be different even for the same TNM staging. The characteristic features of tumors, such as tumor budding, tumor nodules, and extracapsular extension (ECE) of lymph nodes, can influence the disease progression and the outcome for patients. Tumor budding occurs what at the invasion front of colorectal adenocarcinomas, tumor cells, singly or in small aggregates, become detached from the neoplastic glands, and it can be divided it into two groups, low grade (0∼16 foci in a field) and high grade (17 or more foci in a field). A tumor nodule is histologically identified within the fatty tissue or the detached fatty tissue around the dissected lymph nodes, or is a place picked up as lymph nodes from resected specimens which contain no lymph node components. ECE is defined as a tumor extension beyond the node capsule. The aims of this study were to evaluate the clinical significance of tumor budding, tumor nodules, and ECE of lymph nodes as prognostic factors in Stage III colorectal cancer patients. Methods: We analyzed the disease-free and overall 5-year survival rates and recurrence rates in 94 Stage-III colorectal cancer patients according to tumor the budding intensity, the tumor nodules, and the lymph node ECE status. Results: Of the entire group, the 5-year disease-free and overall survival rates were 49%, and 50%, respectively. The 5-year disease-free and overall survival rates were higher in the low-grade tumor budding group than in the high-grade group (58% vs 33%, P=0.045, 61% vs 39%, P=0.003). The 5-year disease-free and overall survival rates in patients with tumor nodules were lower than those in patients without one (44% vs 69%, P=0.086, 47% vs 77%, P=0.018). The recurrence rate was also higher in the group with tumor nodules than without one (80% vs 52%, P=0.045). The 5-year disease-free and overall survival rates were higher in the ECE negative group than in the positive one (68% vs 37%, P=0.018, 75% vs 42%, P=0.001). The recurrence rate was also higher in the ECE positive group than in the negative group (78% vs 46%, P=0.008). The existence of ECE and tumor nodule were strongly related to systemic recurrence (P=0.006, P=0.033), but not to the local recurrence (P=0.777, P=0.611). Considering the analysis of the recurrence pattern by N stage classification, there is no statistical difference in the N2 patient group, but there was in the existence of ECE and tumor nodule were strongly related to the systemic recurrence in N1 group (P=0.019, P=0.028). These three factors were scored according to the existence, and the score range was divided into two prognostic groups, high risk group (≥2) and low risk group (＜2). The high risk group was significantly associated with systemic recurrence (P= 0.004) rather than recurrence (P=0.865), and these score value were only significant in the N1 patient group (P=0.007) rather than in the N2 group (P=0.927). The high risk group also showed poor overall survival rate compared with the low risk one in only the N1 group (P=0.002), but nof in the N2 group (P=0.193). On multivariate analysis, UICC stage and ECE were two significant factors for tumor recurrence and the 5-year disease-free survival rate. Conclusions: These data showed that even if similar lymph node metastasis existed in advanced colorectal cancer patients, there was a different 5-year disease-free survival rate and overall survival rate according to the tumor budding, tumor nodule, and ECE status. On multivariate analysis, UICC stage and ECE were two significant factors for the tumor recurrence and the 5-year disease-free survival rate. Our results suggest that tumor budding, tumor nodule, and ECE of lymph node are excellent parameters to provide a confident prediction of clinical outcome. J Korean Soc Coloproctol 2007;23:460-476