Sentinel lymph nodes in cancer of the oral cavity: is central step-sectioning enough?

Extended histopathologic work-up has increased the detection of micrometastasis in sentinel lymph nodes in malignant melanoma and breast cancer. The aim of this study was to examine if (A) step-sectioning of the central 1000 microM at 250 microM levels with immunostaining were accurate when compared with (B) step-sectioning and immunostaining of the entire sentinel lymph node at 250 microM levels.

Forty patients with T1/T2 cN0 oral cancer were enrolled. Three patients were excluded. In one patient no sentinel lymph node was identified. The remaining two had unidentified sentinel lymph nodes due to lymphoscintigraphic and surgical sampling error. The central 1000 microM of 147 sentinel lymph nodes were step-sectioned in 250-microm intervals and stained with hematoxylin and eosin and CK-KL1. All lymph nodes were recorded as negative or positive for macrometastases or micrometastases. After inclusion of the last patient the residual tissue of the lymph nodes was totally step-sectioned at 250-microm intervals and re-classified. The tumor deposits were divided into macrometastases and micrometastases and ITC.

Method (A) upstaged 17 lymph nodes and 11 patients compared with method (B), which upstaged 22 lymph nodes and 11 patients. Seven of the patients with positive lymph nodes did not change stage. However, four lymph nodes changed from micrometastases to macrometastases. One patient changed from a micrometastasis to four micrometastases. One pN2c patient with bilateral micrometastases did not change stage, but an additional ipsilateral lymph node with a micrometastasis was identified.

Larger tumor deposits and more metastases are identified by more extensive sectioning of the sentinel lymph nodes. None of the patients was false-negative due to histopathologic sampling error, but the results indicate that central step-sectioning of the central 1000 microM cannot completely be relied upon for accurate staging of the patients.