Prognostic value of sentinel lymph node biopsy in the pathologic staging of colorectal cancer patients

BACKGROUND AND OBJECTIVES: Over the last decade, lymphatic mapping and sentinel lymph node (sN) biopsy have greatly increased the possibility of identifying nodal metastasis in clinically node-negative patients with melanoma and breast cancer, thus improving the accuracy of pathologic staging. Recently, sN biopsy has been applied also in colorectal cancer. This prospective study aimed to assess its feasibility and accuracy in predicting regional lymph nodes metastases in colorectal cancer patients as well as the impact on treatment decision-making. MATERIALS AND METHODS: Lymphatic mapping was accomplished by means of blue dye, which was intraoperatively injected into the subserosa overlying the tumor site in 26 patients undergoing colorectal cancer surgery. Following bowel resection, the operative specimen was inspected to identify each blue-stained node, the sN, which was sent separately to the pathologist. One half of each sN was examined by multiple 200 microm sections, while the second half was examined by standard bi-valving technique with hematoxylin-eosin (H and E) staining; all the other regional non-sentinel nodes were routinely examined by standard bi-valving technique and H and E staining. RESULTS: At least one sN was detected in 24 of 26 patients (92.3%); two patients with rectal cancer had no sN identified. Overall, 70 sN were retrieved into the operative specimens, with a mean of 2.9 sNs/patient, and 19 sNs were tumor-positive. An agreement between sN and regional lymph-node status was observed in 20 of 24 patients (83.4%). The sN was histologically negative in two of nine patients with positive regional nodes (sensitivity = 77.8%; false-negative rate of 22.2%); in two of seven patients with tumor-positive sN (28.6%), the sN was the exclusive site of regional nodal metastasis. The negative predictive value was 88.2% (15 of 17 patients), and the overall accuracy was 91.7% (22 of 24 patients). As regards the contribution to the detection of nodal metastasis according to the pathologic technique, standard H and E bi-valving technique detected 16 of 19 tumor-positive sNs (84.2%) while, by means of serial sectioning, metastases were detected in the remaining 3 of 19 sNs (15.8%). CONCLUSIONS: The sN biopsy proved feasible, with a rather short learning curve. The focused analysis of the sN by means of serial sectioning improved the detection rate of nodal metastasis compared to standard bi-valving examination, so that a more accurate nodal staging should be expected; finally, an elective localization of metastasis within the sN was observed in almost one third of regional node-positive patients.