Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
The purpose of this chapter is to present unusual morphologic variants of primary cutaneous melanomas, which are not the subject of specific chapters elsewhere in the book. The main value of mentioning and describing these variants is for the practicing pathologist to be aware of the morphologic spectrum of melanomas. Since there is no known clinical significance or associated molecular signature to the variants described herein, we simply present the microscopic findings and discuss the differential diagnosis.
This variant of melanoma is characterized by distinct cytoplasmic changes: on hematoxylin and eosin (H&E)–stained sections the cytoplasm is not pink (eosinophilic), but “clear.” Sometimes it may be foamy. Often the cytoplasm is enlarged and abundant. Balloon cell changes are most often associated with superficial spreading or nodular melanomas. Although in some melanomas all or nearly all tumor cells display balloon cell features ( Fig. 20.1 ), often those features are focal and part of a melanoma with otherwise conventional cytology.
If a tumor displays nearly exclusively balloon cell features and pagetoid spread is present, it may need to be distinguished from other pagetoid cutaneous clear cell neoplasms, such as sebaceous carcinoma or clear cell sweat gland neoplasms. Nodular balloon cell melanomas may be confused with metastatic renal cell carcinoma or histiocytic tumors, such as a xanthoma.
This variant of melanoma shows some similarity to and may be confused with a deep penetrating nevus. The tumors are typically heavily pigmented and display a plexiform growth pattern. There may be, as in a deep penetrating nevus (DPN), a deeply extending pigmented tumor component ( Fig. 20.2 ). The diagnosis is particularly challenging when a DPN-like melanoma arises in association with a melanocytic nevus, thereby simulating a combined nevus.
DPN-like melanoma can be distinguished from a DPN by its asymmetric silhouette and/or expansile growth, a higher degree and more uniform (as opposed to random scattered) nuclear atypia and higher number of mitoses, including mitoses near the base of the lesion. The presence of associated features of melanoma in situ also facilitates the diagnosis of melanoma. Ancillary studies may also help, as DPN-like melanomas may be associated with molecular aberrations typical of melanoma.
Primary dermal melanoma is listed herein as an unusual form of melanoma. However, it lacks distinctive cytologic features and is in our opinion best regarded as a variant of primary nodular melanoma. By definition, there is a dermal nodule of melanoma with no associated precursor lesion (melanoma in situ or melanocytic nevus) ( Fig. 20.3 ). Accordingly, the main diagnostic challenge is correct staging. Is the dermal melanoma nodule a primary or a metastatic tumor? The initial series of primary dermal melanomas reported a lower Ki-67 labeling index compared with conventional nodular melanomas and suggested a more favorable clinical course. However, this has not been independently validated, and the difference in observed outcomes may have been biased by the selection of tumors with a low proliferative index. Current evidence suggests that primary dermal melanomas are genomically heterogeneous, some carry NRAS and others carry BRAF mutations, whereas some lack both BRAF and NRAS mutations. Some tumors may be associated with gene rearrangements and fusions.
Follicular or folliculotropic melanoma is a variant of melanoma in which the in situ component is predominantly located within folliculosebaceous structures ( Fig. 20.4 ). Melanoma in situ exclusively confined to a follicle (follicular melanoma) is exceedingly rare. More often, there is prominent follicular involvement (folliculotropic melanoma), but a portion of the melanoma in situ tends also to involve adjacent epidermis. Most follicular and folliculotropic melanomas occur in the head and neck region. Associated invasive melanoma is usually found in the perifollicular stroma. Follicular melanomas pose a problem for precise measurement of tumor thickness. Using the granular cell layer of the overlying epidermis as starting point for measuring tumor thickness would often result in an overestimate of tumor thickness. For follicular melanomas, we recommend using the center of the follicle or the epithelial border of the pilar canal closest to the most peripheral invasive melanoma cell as an estimate of tumor thickness. Primary follicular melanoma with no significant in situ component adjacent to the follicle must be distinguished from a folliculotropic melanoma metastasis.
In this variant of melanoma, the tumor cells display “rhabdoid” features characterized by epithelioid cell features and eccentric nuclei with a single prominent nucleolus as well as intracytoplasmic hyaline inclusions ( Fig. 20.5 ). The term rhabdoid refers to cytologic similarity to rhabdoid tumor , a term originally coined by Haas et al. in 1981 to describe an aggressive childhood renal tumor. Tumors with similar features have in the meantime been reported to occur at many other sites (extrarenal rhabdoid tumors).
Rhabdoid melanoma was first reported by Bittesini et al. in 1992. This phenotype is usually seen in metastatic tumors but can also be found in primary tumors. The tumors are usually amelanotic. If there is associated melanoma in situ, the diagnosis is usually straightforward. Otherwise clinical correlation and immunostains may be necessary to distinguish primary nodular melanoma with rhabdoid features from a primary soft tissue or cutaneous metastasis or other malignant neoplasms with rhabdoid features, such as malignant rhabdoid tumor of childhood, malignant peripheral neuroectodermal tumor, or rhabdomyosarcoma.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here