Optimal cut-off value for lymph node harvesting in colon cancer based on tumor stage and sidedness: A retrospective analysis of a Japanese nationwide cohort.

Abstract

149 Background: The presence of fewer than 12 harvested lymph nodes (LN) is currently considered a risk factor in stage II colon cancer (CC). However, this cut-off value does not take into consideration the biological differences associated with tumor sidedness. Moreover, the applicability of this cut-off value to stage III CC is unknown. Methods: Patients with pathological stage II/III CC who underwent curative surgical resection were assessed using data from the Japanese Society for Cancer of the Colon and Rectum database (2001–2015). The optimal cut-off value of harvested LN was evaluated by analyzing all cut-off values (i.e., 3–40), taking into account the point estimate of the hazard ratio (HR) and lower limit of the 95% confidence interval (CI), according to the tumor stage (II/III) and sidedness (right/left). A Cox proportional hazards model was used to investigate the impact of the number of harvested LN on overall survival, adjusting for age, sex, T stage, histology, size of tumor, sidedness, adjuvant chemotherapy and extent of colon resection. Multivariate analysis was also conducted. Results: From the data, 4249 (left/right; 3248/1001) and 3549 (left/right; 2798/751) patients with stage II and III CC, respectively, were identified. The median number of harvested LN was identical between stage II (18, interquartile range [IQR] (12–15)) and III (18, IQR (13–25)). However, this number was higher in right-sided CC (21, IQR (14–31)) versus left-sided CC (17, IQR (12–24)). In stage II, the conventional cut-off value (i.e., 12 LN) was identified as a prognostic factor in the multivariate analysis (HR 1.45; 95% CI: 1.20–1.75; p<0.001). Nevertheless, the prognostic impact of the number of harvested LN was stronger when the optimal cut-off value was set to 7 (HR 1.88; 95% CI: 1.44–2.44; p<0.001). In stage III, the optimal cut-off value was 12 (HR 1.88; 95% CI: 1.44–2.44; p<0.001). When evaluated based on tumor sidedness, the optimal cut-off value differed, especially for stage III. Data are shown in the table. Similar results were obtained following the exclusion of the extent of bowel resection as variable in Multivariate analysis. Conclusions: While the presence of 12 harvested LNs remains a valid prognostic factor, the optimal cut-off value can be further evaluated based on tumor stage and sidedness. Stage II Stage III All Left Right All Left Right Cut-off value 12 12 12 12 12 12 N(>cut-off/<cut-off) 3209/1040 2371/877 838/163 2814/735 2181/617 633/118 HR(>cut-off/<cut-off)(95% CI) 1.45 (1.20-1.75) 1.49 (1.20-1.84) 1.31 (0.85-2.02) 1.30 (1.09-1.55) 1.27 (1.05-1.54) 1.43 (0.97-2.13) P value <0.001 <0.001 0.216 <0.003 0.015 0.073 Cut-off value 7 7 10 20 13 N(>cut-off/<cut off) 3930/319 2977/271 889/112 1200/1598 609/142 HR(>cut-off/<cut-off)(95% CI) 1.88 (1.44-2.44) 2.07 (1.56-2.74) 1.39 (0.84-2.27) 1.35 (1.13-1.60) 1.59 (1.11-2.28) P value <0.001 <0.001 0.198 <0.001 0.011