[Prognostic value and initial exploratory research on TNM staging method of tumor deposits in stage III colon cancer]

Objective: To investigate the effect of tumor deposits (TD) on the prognosis of patients with stage III colon cancer, and to explore whether TD number included into regional lymph node count can predict the prognosis more accurately. Methods: A retrospective cohort study was carried out. Case inclusion criteria: (1) primary colon cancer; (2) undergoing colon cancer radical operation; (3) definite pathological diagnosis; (4) colon cancer stage III according to AJCC 8th edition; (5) complete follow-up data; (6) without preoperative neoadjuvant treatment. Clinicopathological data of 296 patients undergoing colon cancer radical operation from January 2005 to December 2008 in the Cancer Hospital of Chinese Academy of Medical Sciences were retrospectively collected. The effect of TD and its amount on the prognosis was evaluated. Colon cancer TNM staging method based on the 8th edition of AJCC was compared with the modified TNM staging (mTNM) adjusted by the number of TD. The differences of the disease-free survival (DFS) and overall survival (OS) between groups were also examined. The Kaplan-Meier curve was used to analyze the survival, and prognostic factors were analyzed by Cox univariate and multivariate analyses. Results: Among 296 patients with stage III colon cancer, 78 patients had TD. The median number of TD was 2 (1-10). Tumor T stage, N stage, vascular tumor thrombus and preoperative carcinoembryonic antigen (CEA) were associated with TD in patients with colon cancer (all P<0.05). The right hemicolon appears likely to have TD than left hemicolon, but the difference was not statistically significant (P=0.059). The median follow-up of the whole group was 71 (6-102) months. During the follow-up period, 129 patients (43.6%) had recurrence or metastasis, and 111 patients died (37.5%). The 5-year DFS in TD group was 44.9%, which was lower than that in the non-TD group (60.6%), with statistically significant difference (P=0.003). The 5-year OS in TD group was 50.0%, which was also lower than 67.0% in the non-TD group, and the difference was statistically significant (P=0.002). According to TD number, patients were divided into 3 groups: 1 TD (25 cases), 2-3 TD (32 cases), ≥4 TD (21 cases). The 5-year DFS in these 3 groups was 68%, 56.3%, and 0, respectively (P<0.001), and 5-year OS was 76%, 59.4%, and 4.8% respectively (P<0.001). Univariate analysis showed that TD presence (95% CI: 1.234-2.694, P=0.003) and TD number (95% CI: 3.531-14.138, P<0.001) were associated with the prognosis of patients with stage III colon cancer. At the same time, age, tumor N stage, tumor location, chemotherapy, and preoperative CEA elevation were also associated with the prognosis of stage III colon cancer patients (all P<0.05). Multivariate analysis revealed that TD presence (HR=1.957, 95%CI: 1.269-3.017, P=0.002) and TD number (HR=8.020, 95% CI: 3.414-18.842, P<0.001) were still independent risk factors for the prognosis of patients with stage III colon cancer.According to the TD number counted as metastatic lymph nodes, in 78 patients with TD, 24 patients were upstaged in N stage, and 16 patients upstaged from TNM stage IIIB to stage IIIC. For 16 stage IIIB cases with staging modification, 30 unadjusted stage IIIB cases with TD, and 148 stage IIIB cases without TD, the 5-year OS was 37.5%, 73.3% and 76.4%, respectively with significant difference (P<0.001). However, for 16 patients adjusted as stage IIIC (mTNM), 32 patients with unchanged stage IIIC with TD (TNM, AJCC 8th edition), and 63 stage IIIC cases without TD, the 5-year OS was 37.5%, 36.4%, and 41.3%, respectively without significant difference (P=0.707). Conclusions: TD presence and TD number are independent risk factors for prognosis of stage III colon cancerpatients. TNM staging evaluation with lymph node number including TD number can predict the prognosis of patients more accurately.