The Number of Lymph Nodes Examined Debate in Colon Cancer: How Much Is Enough?

Abstract

Background: Much debate exists over the significance of the number of lymph nodes (LN) examined after colon resection for colonic adenocarcinoma. We wished to use a population-based dataset to help clarify the optimal number of LN to examine. Methods: The Surveillance, Epidemiology and End Results (SEER 1988-2005) database was queried for patients who presented with colonic adenocarcinoma and had data recorded for the number of LN examined. Patients who had zero LN examined or had a survival of less than one month were excluded. Multiple Cox proportional hazard regressions were run using successive LN examined cutoffs (5-25), first controlling for and then stratifying by T-stage of the patient's tumor. Additional variables controlled for in every regression were: age, gender, ethnicity, marital status, number of positive LN, grade, presence of metastatic lesions at diagnosis and extent of primary surgical procedure. After each regression, a Harrell's C Statistic and an Akaike's Information Criterion (AIC) was run to test the predictive capacity and fit of the model, respectively. Where two models in each comparable population had equally high Harrell's C Statistic's, the one with the lower AIC was chosen as the best model. Results: 128,071 patients met selection criteria. Median age was 70 years and median survival was 83 months. Table 1 shows the hazard ratio (HR), 95%CI, P value, Harrell's C Statistic and AIC for each model while controlling for T stage. While the HR for each integer increase steadily decreased, the Harrell's C Statistic shows that the cutoffs at 14 LN and 15 LN were the most predictive. Between those, the AIC shows that the cutoff at 15 LN fit the data more closely than the 14 LN cutoff. This process was repeated for subsets stratifying by T stage. The models with the best predictive ability and best fit by T stage were: T1, 14 LN (HR=0.89, 95%CI=0.79-1.00, p=0.050, Harrell's C=0.7322, AIC=33247.31); T2, 10 LN (HR=0.87, 95%CI=0.82-0.93, p<0.001, Harrell's C=0.7105, AIC=68369.96); T3, 10 LN (HR=0.77, 95%CI=0.75-0.79, p<0.001, Harrell's C=0.7105, AIC=478225); T4, 12 LN (HR=0.73, 95%CI=0.70-.077, p<0.001, Harrell's C=0.7101, AIC=128725). Conclusions: Using a population-based dataset, we show the optimal number of LN examined is dependent upon the patients tumor stage. Across all T stages the highest optimal and that patients with 15. Since is possible to estimate but not perfectly predict the T stage of a patient's tumor pre-operatively, we believe the recommendation should be based on the most conservative measure.