Are deeper sections and immunohistochemistry useful in detecting micrometastases and isolated tumour cells in colorectal cancer?

Identification of lymph node metastases is critical for staging of colorectal cancer. Lymph node metastases are classified based on size as macrometastases, micrometastases, or isolated tumour cells (ITCs). Micrometastases are associated with worse prognosis; however, optimal detection methods have yet to be established. The first objective was to determine if deeper levels and immunohistochemistry would detect micrometastases in patients with metastatic disease but negative lymph nodes. Five patients with pT3N0 colorectal adenocarcinoma who developed metastatic disease were identified. Three deeper haematoxylin and eosin (H&E) levels followed by pan-cytokeratin immunohistochemistry was performed on all lymph nodes. No micrometastases were detected; however, ITCs were seen by immunohistochemistry in three of five patients. Driven by these findings, the second objective was to determine if a single level stained for pan-cytokeratin would identify ITCs and if their presence was associated with an increased risk of disease recurrence. A cohort of eight patients with stage IIA (pT3N0M0) colorectal adenocarcinoma who developed distant metastasis was matched to eight control patients who remained disease-free over a 5-year period, and a single pan-cytokeratin stain was performed on all lymph nodes. ITCs were identified in six of eight patients that developed metastasis and in five of eight control patients (odds ratio=1.80; 95% confidence interval=0.21-15.41). In conclusion, three deeper levels and immunohistochemistry did not increase the yield of micrometastases in pT3N0 colorectal adenocarcinoma. While ITCs were readily identified by immunohistochemistry, their presence was not a significant predictor of distant recurrence. These findings do not support the routine use of deeper levels and immunohistochemistry for lymph node staging.