Ex vivo sentinel lymph node mapping in colon cancer: improving the accuracy of pathologic staging?

Background A subset of patients with colon cancer staged by conventional methods have occult micrometastases and do not receive adjuvant chemotherapy. Sentinel lymph node (SLN) mapping and staining by immunohistochemistry is a technique that may identify such occult micrometastases, thereby upstaging patients with positive findings. The purpose of this study was to determine whether ex vivo SLN mapping in colon cancer could be applied successfully to patients at our institution. Methods Seventeen patients with intraperitoneal colon tumors undergoing resection were studied prospectively. SLNs were identified as the first blue stained node(s) after ex vivo peritumoral injection of isosulfan blue dye. Additional lymph nodes were harvested and processed in accordance with standard pathologic evaluation for colon cancer. All nodes were examined after routine hematoxylin and eosin (H&E) staining. SLNs that were negative on H&E were analyzed further by multilevel sectioning and immunohistochemistry staining using anticytokeratin monoclonal antibody. Results Of the 17 study patients, SLNs were identified in 16 (94%) cases. The SLN was the only positive node in 3 patients. An identified SLN was positive (by H&E) in all patients with associated positive non-SLN nodes. The average number of nodes retrieved per patient was 16 (range, 4–54). Overall, SLNs accurately reflected the status of the entire lymph node basin in 16 (94%) patients. Two (12%) patients with negative nodes by H&E potentially were upstaged after further SLN analysis. The negative predictive value for SLN mapping was 89%. Conclusions The ex vivo technique of SLN mapping for colon cancer is feasible. In the current study, SLN results were concordant with non-SLNs in the majority of patients. Furthermore, this technique may have upstaged 2 (12%) patients. Whether this ultimately will affect overall survival has yet to be determined.