Annals of Surgical Oncology

, Volume 7, Issue 8, pp 601–608

Lymph node micrometastases do not predict relapse in stage II colon cancer


  • Jorg Tschmelitsch
    • Colorectal ServiceMemorial Sloan-Kettering Cancer Center
  • David S. Klimstra
    • From the Department of PathologyMemorial Sloan-Kettering Cancer Center
    • Colorectal ServiceMemorial Sloan-Kettering Cancer Center
Original Articles

DOI: 10.1007/BF02725340

Cite this article as:
Tschmelitsch, J., Klimstra, D.S. & Cohen, A.M. Ann Surg Oncol (2000) 7: 601. doi:10.1007/BF02725340


Background: Over one third of patients with stage II colonic adenocarcinoma experience tumor recurrence. Because effective adjuvant therapy is now available, it is important to identify subsets of patients at higher risk for relapse who may benefit from early treatment. Immunohistochemistry has been used to detect microscopic metastases in histologically uninvolved mesenteric lymph nodes, but the prognostic significance of minimal nodal involvement has not been established.

Methods: Hematoxylin and eosin (H&E)-stained recuts of 900 mesenteric lymph nodes from 55 patients (range, 2–47; mean, 16.4 nodes per case) with resected pT3 or pT4, N0, M0 (TNM stage II) colonic adenocarcinomas were re-examined for the presence of metastases and then stained immunohistochemically for keratin using the AE1:AE3 antibody. Twenty-seven patients did not experience recurrence of tumor within 5 years following resection (no evidence of disease [NED]); 28 patients relapsed during the same time frame. Lymph nodes from 10 patients having colonic resections for nonneoplastic disorders also were stained as controls. Keratin-positive cells and cell clusters were quantified in the lymph nodes, and comparisons were made between patients with and without tumor relapse.

Results: In the relapse group, four patients had positive nodes already identified on the H&E-stained recuts and had to be excluded from further analysis. Sixteen additional patients had keratin-positive cells; thus, 16 of 24 (67%) had micrometastases. In the NED group, one patient had a positive node on H&E staining and 22 additional patients had keratin-positive cells, so 22 of 26 (84%) patients had micrometastases. In the patients who had micrometastases, there was a mean of 3.5 and 4.6 positive nodes in the relapse and NED groups, respectively, and a mean of 11.3 and 12.4 keratin-positive cells or clusters in the relapse and NED groups, respectively. No keratin-positive cells were found in the 1 to 21 (mean, 9.1) nodes per case studied in the control patients.

Conclusions: Micrometastases to histologically uninvolved mesenteric lymph nodes commonly are detected in patients with pT3 or pT4 colonic adenocarcinomas on recuts stained immunohistochemically for keratin. Nodal micrometastases detected by immunohistochemical staining are not useful for identifying stage II patients at higher risk for relapse.

Key Word

Colonic adenocarcinomaMicrometastasesImmunohistochemistryKeratin
Download to read the full article text

Copyright information

© The Society of Surgical Oncology, Inc 2000