Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Most patients with primary cutaneous melanoma will be cured by surgery. However, metastatic disease develops in approximately 15% to 30% of patients. Most metastases manifest after the diagnosis of the primary melanoma, usually within less than 3 years, rarely up to decades later. On occasion the metastatic disease may be apparent at the same time when the primary tumor is found or present in the absence of a known or detectable primary melanoma. Tumor cells can separate and spread from the primary melanoma via lymphatics or hematogenously (as tumor embolus or in an extravascular migratory pattern) or along neurovascular structures.
Metastatic melanoma may come to clinical attention because of symptoms or physical signs, with specific manifestations depending in large part on the location and volume of the tumor(s) (e.g., palpable lymph node, soft tissue or skin nodule(s), headaches or seizure related to brain metastasis, bowel obstruction from intestinal metastasis, bone pain from bone metastasis, general fatigue, and weight loss due to disseminated disease). Metastases anywhere may be small or large, solitary or multiple.
At the present time, the first metastasis in a patient with known primary melanoma is often detected in a clinically asymptomatic individual by sentinel lymph node (SLN) biopsy (SLNB), clinical examination, or radiologic imaging during initial staging workup or on a follow-up visit. Suspected metastatic lesions usually require a tissue biopsy for histopathologic confirmation of the diagnosis and to assess the effects of modern systemic treatment options dependent on findings in the tumor or its microenvironment.
Cutaneous metastases deserve special mention because they pose distinct clinical diagnostic challenges: not only does metastatic melanoma to the skin need to be distinguished from miscellaneous nonmelanocytic tumors, but they can also be confused with a primary melanoma or even a melanocytic nevus. Sometimes the diagnosis of metastatic melanoma may be immediately obvious when multiple new pigmented papules or nodules surround the primary tumor ( Fig. 27.1 ) or a metastasis develops in a patient with a known history of melanoma. However, an isolated amelanotic papule nodule may be clinically difficult or impossible to diagnose, especially in the absence of a history of melanoma. The clinical appearance of individual lesions may strongly suggest melanoma (e.g., ulcerated pigmented nodule) or be subtle and nonspecific. Dermoscopic features associated with an amelanotic cutaneous melanoma metastasis include an abnormal vascular pattern ( Fig. 27.2 ).
The principle pathologic finding of metastatic melanoma is the detection of melanoma cells at a site and tissue environment away from the primary tumor ( Fig. 27.3 ). Metastatic lesions are generally classified into locoregional and distant tumor deposits.
Locoregional metastases are metastatic tumor deposits in the regional lymph node basin draining the primary tumor site and/or in skin or soft tissue located between the primary tumor and lymph node basin. In the American Joint Committee on Cancer (AJCC) staging system, locoregional metastases represent stage III disease.
A lymph node may be removed because of clinical suspicion (e.g., palpable enlarged node) or as part of an SLNB. Melanoma may be found anywhere in the node. Tumor deposits are typically found in the subcapsular sinus and periphery of the node ( Fig. 27.4 ). Under the microscope, tumor cells appear “foreign” to the node (i.e., they look different from the background of normal constituents of the lymph node). Larger cell aggregates tend to distort the architecture of the node and are thereby more easily recognized ( Fig. 27.5 ). There is great variability in the cytologic appearance of metastatic melanoma, ranging from round to oval to fusiform, small and large, pigmented and nonpigmented cytoplasm ( Box 27.1 ). The diagnosis of a pigmented melanoma (tumor cells with cytoplasmic melanin pigment) in a lymph node is usually straightforward. Amelanotic malignant spindle and epithelioid cell deposits are also often readily identified and confidently diagnosed as metastatic melanoma in the right clinical setting (known history of invasive melanoma, primary tumor available for comparison). However, immunohistochemical stains may be necessary or prudent to detect or confirm the diagnosis of metastatic melanoma (see Fig. 27.5 ), especially if the patient's primary tumor is not available for review, the patient has a history of other cancers, and/or the microscopic features of the tumor suggest a differential diagnosis beyond metastatic melanoma. In this regard, the morphologic features of the suspicious cells should be taken into account when selecting the markers of melanocytic differentiation to be used (i.e., for possible spindle cell melanoma metastasis IHC for Sox10 and/or S100 are critical because other melanocytic markers may be negative). When metastatic melanoma has been excluded or deemed unlikely based on the immunophenotype, the microscopic findings and clinical setting should guide the choice of additional markers to establish the correct alternate diagnosis. For example, if an undifferentiated epithelioid malignant neoplasm is found in an axillary node from a patient with a history of breast cancer and melanoma, the panel of markers should include miscellaneous cytokeratins, EMA, and estrogen receptors.
Melanoma in unexpected sites (e.g., subcutis, lymph nodes, viscera)
Nodular, circumscribed, or infiltrative
Necrosis common
Nests, sheets, fascicles, or dispersed single cells
Epithelioid and/or spindle cells
Enlarged nuclei
Nuclear pseudoinclusions
Large eosinophilic or amphophilic nucleoli and/or hyperchromatism
Mitotic figures common
Moderately abundant pale eosinophilic cytoplasm
Melanin pigment (intracytoplasmic) may be present or absent
Sparse lymphocytic infiltrate (usually peritumoral)
Minimal fibrosis (usually absent)
Melanin-laden macrophages, often associated with pigmented melanoma or necrosis
In the contemporary clinical management of patients with primary cutaneous melanoma, stage III disease is often detected as part of an SLNB. This procedure is routinely performed at many melanoma treatment centers for clinically node-negative patients with primary tumors categorized according to AJCC criteria as pT1b or higher because the SLN status is a strong predictor of outcome. The 5-year survival of SLN-negative patients is 90% compared with approximately 70% for those in whom the SLNB is positive. Metastases are identified in 15% to 20% of melanoma patients who undergo an SLNB. Although SLNB is currently accepted as a meaningful staging procedure for prognosis, recently reported phase III randomized clinical trials demonstrated no improvement in overall survival when a completion lymph node dissection (CLND) was performed in patients with a positive SLN. However, the status of the non-SLN has prognostic value, and in a select group of patients CLND may allow better control of regional disease.
The SLN is defined as the lymph node that receives direct lymphatic drainage from the primary tumor site. Although most laboratories follow similar principles in the diagnostic workup of an SLN, there is no established consensus among pathologists on the best method for pathologic examination. Most pathologists agree that an SLN should be initially bisected along its longitudinal axis. For larger nodes (>10 mm in thickness), additional parallel slices may need to be taken. Procedures between laboratories vary the most in the number of sections and immunostains that are taken from each tissue block. In our opinion, pathologists should examine each tissue slice of an SLN by at least one H&E and one or two immunostains for a melanocyte differentiation antigen, such as Melan-A/Mart1 and/or HMB-45. If the primary tumor is known to be negative for these markers (e.g., amelanotic spindle cell melanoma), it is important that immunohistochemistry (IHC) for S100 protein and/or Sox10. In diagnostically difficult cases (e.g., nodal nevus vs. metastatic melanoma, history of other malignancy that could involve the SLN), more stains may be needed. Although using more sections and immunostains in the analysis of SLNs will yield a higher number of “positive” cases, we do not recommend exhausting the available tissue. First, the clinical significance of some “positive” SLNs, such as when only one or a few rare immunoreactive cells are identified only by extensive sectioning protocols using numerous immunostains, is uncertain. Second, it is simply impractical to analyze lymph node tissue comprehensively in search for a rare isolated tumor cell. Nevertheless, because the SLN status is important prognostically for individuals with localized primary melanoma, pathologists are obliged to make a diligent assessment of it.
Several studies have demonstrated that the location and extent of tumor deposits within an SLN has predictive value for the probability of positive non-SLNs and implicitly for prognosis. In other words, finding only a few melanoma cells in the subcapsular sinus of an SLN is prognostically not as unfavorable as multiple large tumor deposits, especially when they replace much of the node and/or extend into extranodal soft tissue.
Microscopic features of melanoma deposits in the SLN that have been proposed to be prognostically informative and are therefore worth reporting include the size of metastases, maximum subcapsular depth (centripetal thickness or tumor penetrative depth), the microanatomic location of SLN tumor deposits, and the presence of extracapsular spread. The determination of some of these parameters may at times be difficult and not reliable, such as when tumor deposits are irregularly shaped or widely dispersed throughout the node as a mix of small aggregates and solitary tumor cells ( Fig. 27.6 ). Furthermore, the tumor burden is to some degree dependent on the extent of sectioning (more sections may reveal additional tumor deposits or demonstrate a greater dimension of deposit(s) in the deeper sections). Nonetheless, often a reasonable attempt can be made for a semiquantitative assessment of tumor burden.
Locoregional metastases to the skin or soft tissue have historically been termed “satellite” or “in-transit” metastasis, depending on their distance from the primary tumor. Satellites have been distinguished into grossly visible lesions and microscopic satellites. The latter are defined as metastatic tumor cells microscopically seen away from the primary tumor (usually referring to a biopsy or excision that contains both primary tumor or a scar and separate metastatic deposits). In-transit metastases have been defined clinically on the basis of their proximity to the primary tumor site (with arbitrary cutoffs of 2 or 5 cm most commonly applied). The presence of locoregional metastases is prognostically unfavorable. As long as the tumor is truly a locoregional metastatic deposit (and not part of the primary tumor seemingly separated by a scar or regression or regrowth of an incompletely removed primary), the precise distance from the primary tumor and location between primary and SLN does not matter for prognostic purposes.
Locoregional soft tissue or skin metastases may manifest as solitary tumor or multiple lesions (see Fig. 27.1 ). The size of a tumor nodule is variable. Soft tissue metastases tend to be larger at the time of biopsy than most dermal metastases.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here