Racial disparities in lymph node dissection for colon cancer

A68

Introduction: Blacks have higher colon cancer incidence and mortality rates, are more likely to be diagnosed at an advanced stage, and have worse stage-for-stage survival when compared to whites. The number of lymph nodes (LN) examined has been shown to be an important prognostic factor for colon cancer. We hypothesized that differences in the extent of LN dissection contribute to the observed survival disparity between blacks and whites.
 Methods: We selected AJCC stage 1-3 colon cancer patients identified as black or white in the Surveillance, Epidemiology, and End Results database from 1988-2003. Patients were excluded if they had distant metastases, were not histologically confirmed, had an unspecified number of LN sampled, or if they had sarcomas or lymphomas of the colon. We compared demographic and clinical characteristics by race. We developed logistic regression models of the association of race with nodal status, and the ratio of positive to examined LN. Cox proportional hazards models were used to evaluate the association between race and mortality, adjusting for the other variables.
 Results: Among 110,529 cases of stage 1-3 colon cancer (11,038 black and 99,491 white), the proportions of blacks vs. whites who receive nodal dissections according to the ASCO guidelines was significantly higher in blacks compared to whites (45.1% vs. 43.6%, p<0.0001). For tumor stages 1 and 2, significantly higher proportions of blacks vs. whites had ≥12 nodes sampled (p<0.0001). In tumor stages 3 and 4, these proportions were statistically equivalent. Similarly, across all node stages, the proportions of blacks vs. whites who receive nodal dissections in keeping with the ASCO guidelines are either equivalent or marginally higher among blacks. Across all tumor stages, blacks have a significantly greater odds of having 1 or more positive nodes compared to whites (OR=1.27, 95%CI 1.22-1.33). This association is only marginally reduced after adjustment for age, sex, tumor grade, and nodes examined (OR=1.25, 95%CI 1.12-1.30). The extent of LN sampling is an important independent prognostic factor for the overall survival of colon cancer. We report a 20% improvement in overall survival (HR= 0.79, 95%CI 0.78-0.81) among patients who had ≥12 vs. 1-12 nodes sampled. However, since Blacks and whites do not differ in regard to extent of LN sampling, additional adjustment for extent of the LN dissection had no effect on our estimate of the racial survival disparities.
 Conclusions: Based on our dataset, we found that blacks and whites receive equivalent quality of care in regard to the extent of LN sampling. We conclude that the racial survival disparity is not due to differences in quality of surgery or surgical staging. Across all tumor stage categories, blacks were more likely to have positive LN, even after adjustment for the extent of LN sampling, age, sex, grade, and SES. The poorer survival experienced by blacks with stage 1-3 colon cancer cannot be explained by less extensive LN dissection. Future efforts should focus on the elimination of potential differences in non-surgical treatments, such as chemotherapy and supportive care. Our finding that blacks have a consistently higher likelihood of nodal involvement, regardless of tumor stage, suggests that there may also be etiologic differences which predispose this population to more invasive cancers which may be more challenging to treat.