Factors influencing the number of retrieved lymph nodes (LNs) in patients with stage II colon cancer: LN size and areas of B cells and T cells.

Abstract

577 Background: In patients with stage II colon cancer, a high number of retrieved LNs has been reported to be associated with better outcomes. We examined the association between a high number of retrieved LNs and the presence of large mesenteric LNs and we evaluated the areas occupied by the T-cell and B-cell fields in LNs with large diameters. Methods: The study group comprised 320 patients with stage II colon cancer who underwent R0 resection from 1991 to 2003. The long- and short-axis diameters of 4,745 LNs were measured with a digitizer on specimens stained with hematoxylin and eosin. A LN with the largest long-axis diameter was selected for each patient and stained immunohistochemically using CD3 as a marker of T cells and CD20 as a marker of B cells. The CD3-positive and CD20-positive area ratios were evaluated, respectively. Results: The mean number of retrieved LNs was 15±10 per case. The correlation coefficients between the numbers and the long-axis diameters of LNs were as follows: mean 0.23, median 0.16, maximum 0.59, and minimum -0.29. The correlation coefficients between the numbers and short-axis diameters were 0.18, 0.13, 0.54, and -0.33, respectively. The correlation coefficient was highest for the maximum long-axis diameter. When the maximum diameter was divided at 5-mm intervals, the number of LNs significantly increased in parallel to an increase in the maximum long-axis diameter (p<0.001). The median follow-up was 119 months. Fifty-six patients (18%) had recurrence, 110 (34%) died of all causes, and 48 (15%) died of colon cancer. On survival analysis, age, T stage, and maximum long axis diameter of LNs were independent prognostic factors. The CD20 area ratio was significantly higher in LNs with a long-axis diameter of 10 mm or greater (0.44±0.09) than in LNs with a long-axis diameter of less than 10 mm (0.41±0.10, p=0.018). There were no differences in the CD3 area ratio. Conclusions: In stage II colon cancer, the number of retrieved LNs correlated with the maximum long-axis diameter of LNs. The long-axis diameter as well as age and T stage were independent prognostic factors. LNs with a larger long-axis diameter had a significantly increased intranodal B-cell area ratio.