Nevus Cell Aggregates in Lymph Nodes

Abstract

The article by Holt et al1 in this issue of the Journal looks again at the issue of nevus cell aggregates in lymph nodes and contributes substantially to our understanding of the phenomenon. At the same time, like all good studies, it raises some important questions that remain to be answered. In this editorial, I briefly summarize what is known about the frequency of nevus cells in lymph nodes, their locations in terms of gross and microscopic anatomy, the means used to identify them, and the prevailing theories of how nevus cells might get there in the first place. Finally, I briefly mention some of the practical diagnostic dilemmas that can arise when nevus cells are found in lymph nodes. The occurrence of nevus cell aggregates in lymph nodes has been known at least since Stewart and Copeland’s 1931 report, and subsequent studies indicate that it no longer can be considered a rare phenomenon.2 Among patients with melanoma undergoing lymph node biopsy or lymphadenectomy, 3% to 22% have had nodal nevi.3,4 In patients with melanoma, nodal nevi have been found in 0.12% to 0.54% of nodes from full lymph node dissections,3,5 1.2% of nodes from selective lymph node dissections,4 and 3.9% to 13% of sentinel lymph nodes.4,6 Their frequency is less in lymphadenectomy specimens for breast carcinoma, having been found in 0.33% of patients and in 0.017% to 0.1% of lymph nodes in this setting.3,4 Nevus cells also have been encountered in lymph nodes from patients with no known malignancy, as well as in regional nodes associated with adnexal carcinoma of the skin and squamous cell carcinoma of the tonsil.7 The greater frequency of nevus cell aggregates in sentinel nodes removed for melanoma, as opposed to nonsentinel nodes, has been demonstrated convincingly, although the search for melanocytes in these sentinel nodes tends to be more vigorous, often including multiple levels and immunohistochemical analysis. Nevus cells have been found most often in axillary lymph nodes, but they also have been identified in cervical and inguinal regions; involvement of visceral nodes is rare.4,8 In most cases, the cellular aggregates have the appearance of conventional nevus cells. However, node involvement also has been seen with blue nevus, cellular blue nevus, and plexiform spindle cell nevus, as well as atypical spitzoid tumors.3,9-12 The most common locations of nevus cell aggregates, by far, are the capsule or trabeculae of lymph nodes.3,4,6 Occasionally they can be observed adjacent to small vessels4,9 or within lymphatic vessels surrounding the nodes.13 The nevus cells are surrounded by a delicate reticulin meshwork, a finding not associated with melanoma cells.14 However, there have been a number of reports of nevus cells within cortical or medullary parenchyma or in marginal sinuses3,13,15; Biddle et al7 recently reported 13 such cases. When present in nodes, cellular blue nevi often are found in the peripheral sinuses or parenchyma,16,17 a finding that has led some to postulate a different mechanism from the usual capsular site of conventional nevus cell aggregates.3,18 Nevus cell aggregates most often are found on H&E-stained sections, but additional cases are identified with immunohistochemical analysis.19 The cells most often are positive for S-100 or MART-1, but they are negative or only weakly positive for HMB-45.4,7 In contrast with metastatic melanoma cells, nevus cell aggregates tend to be negative for the proliferation marker Ki-67.7,20 The theories of how nevus cells arrive in or around lymph nodes can be divided into 2 broad categories: (1) arrested migration of neural crest progenitor cells during