Importance of tumor load in the sentinel node in melanoma: clinical dilemmas

Authors:

Abstract
Sentinel lymph-node biopsy is an accurate staging procedure for patients with melanoma. However, this procedure followed by immediate completion lymph-node dissection (CLND) does not seem to improve survival, based on data from the MSLT-1 randomized trial. Patients with micrometastases <0.1 mm have a similar clinical course to sentinel-node-negative patients and routine CLND may not be indicated in these patients. The authors discuss the issue of prognostic false positivity and explain how the Rotterdam Criteria could be useful for the measurement of sentinel node tumor burden. There are two hypotheses to explain melanoma dissemination: first, simultaneous lymphatic and hematogeneous spread, with regional lymph nodes as indicators of metastatic disease; and second, orderly progression, with regional lymph nodes as governors of metastatic disease. The sentinel node (SN) has been defined as the first draining lymph node from a tumor and is harvested with the use of the triple technique and is processed by an extensive pathology protocol. The SN status is a strong prognostic factor for survival (83–94% for SN negative, 56–75% SN-positive patients). False-negative rates are considerable (9–21%). Preliminary results of the MSLT-1 trial did not demonstrate a survival benefit for the SN procedure, although a subgroup analysis indicates a possible benefit. A mathematical model has demonstrated 24% prognostic false positivity. SN tumor burden represents a heterogenous patient population and is classified most frequently with the Starz, Dewar or Rotterdam Criteria. A completion lymph-node dissection might not be indicated in all SN-positive patients. Patients classified with metastases <0.1 mm by the Rotterdam Criteria have excellent survival rates. Ultrasound-guided fine-needle aspiration cytology is emerging as a staging tool for high-risk patients, but more research is necessary before this can change clinical practice.

This publication has 87 references indexed in Scilit: