Introduction

Although melanoma represents only 10% of skin cancers, it accounts for at least 65% of skin cancer-related deaths. Because current systemic therapy fails to offer durable complete response, surgical intervention remains central to the treatment of melanoma. Optimizing surgical interventions rests upon accurate histologic assessment of the primary tumor and staging. To this end, the American Joint Committee on Cancer (AJCC) Melanoma Staging Committee, a multinational collaboration of melanoma researchers from North America, Australia, and Europe, has worked to pool staging and outcome data to provide staging and treatment recommendations. The most current AJCC Cancer Staging Manual, published in 2009, demonstrated that the strongest prognostic indicators of survival are tumor thickness (Breslow depth), mitotic rate, and the presence of ulceration in the primary tumor. Though improvements in disease-free survival have been achieved with various surgical and systemic interventions, significant improvement in survival has not been consistently achieved. As a result, there is continued open discourse regarding the utility of certain treatment practices in the contexts of disease-free survival and overall survival.

Histologic Subtypes of Melanoma

Although the pathogenesis is unclear, melanoma is commonly thought to arise from epidermal melanocytes that have undergone oncogenic transformation. One widely accepted model for growth describes an initial, intraepidermal growth phase followed by a more aggressive vertical growth phase. Cutaneous melanomas may present histologically as a number of different subtypes, with the most prevalent summarized by the AJCC as follows: superficial spreading, lentigo maligna, nodular, acral-lentiginous, desmoplastic, childhood, or unclassified types. Melanomas can arise in the background of preexisting nevi, as described in up to 25% of cases, although most commonly melanomas arise de novo.

Shared by all histologic subtypes of melanoma is the malignant melanocyte. These cells can present as round (epithelioid) or flat (spindled) cells with a prominent nucleus, occasional nuclear pseudo-inclusions, cytologic atypia, mitoses, and blue-gray cytoplasm on routine hematoxylin and eosin staining. Melanomas tend to display poor nesting, with single cells predominating over nested architecture as it invades into the surrounding tissue. The use of immunostaining has been critical in the separation of melanomas from their benign counterparts, the nevomelanocellular nevus, and the histologically atypical but benign possible precursor lesion, the dysplastic nevus. These immunostains have included S-100, MART-1 (Melan-A), and HMB-45. Additional immunostains, including tyrosinase and Ki67 (MIB-1), have also been crucial in the diagnosis of melanoma. These immunostains have also been included in the most recent AJCC as an acceptable tool in the detection of microscopic nodal disease. Unfortunately, however, there is no one stain that defines malignant melanocyte behavior. The diagnosis of melanomas, as a result, is an art best left to skilled dermatopathologists and pathologists experienced in the assessment of melanocytic lesions.

Superficial spreading melanoma, the most common subtype, appears classically on the thigh of a woman or on the back of a man. Lentigo maligna melanoma, the next most common subtype, tends to arise on chronically sun-damaged skin. It is worth mentioning that the nomenclature for melanomas confined to the epidermis, the in situ melanomas, is confusing. These non-invasive tumors are superficial and, as a result, are not given a Breslow depth. At times, melanomas in situ arising in chronically sun-damaged skin, commonly in individuals of advanced age, are referred to as “lentigo malignas” without explicit mention of the words “melanoma in situ.” This can lead to confusion, however, because there exists a lentigo maligna melanoma, an invasive tumor with an intraepidermal component of lentigo maligna overlying invasive disease. There also exists a subtype of melanoma, “superficial spreading melanoma in situ,” which may also be termed simply “melanoma in situ.” In ambiguous cases, the pathologist can provide critical information as to the nature of the tumor. Nodular melanomas are melanomas that tend to be identified during an expansive vertical-growth phase early in the life of the melanoma. In people of Asian and African descent, acral-lentiginous melanomas predominate, occurring on the hands and feet. Desmoplastic melanomas, although they represent only 1–4% of the overall melanomas, are notoriously difficult to diagnose clinically and histologically. Clinically, they can be amelanotic, presenting only as an innocuous, erythematous papule or patch. Histologically, this lesion is characterized by a population of spindled melanocytes within a fibrotic stroma resembling a scar and demonstrates an unusual staining profile with negative staining for the most sensitive melanoma markers (MART-1 and HMB-45).

There have been efforts to clarify a group of melanocytic tumors of unclear malignant potential (MELTUMP) that plague the dermatopathology world. These tumors encompass the precursor and borderline lesions that may have features of both benignity and malignancy. One of the classic atypical melanocytic lesions is the so-called benign juvenile melanoma, the Spitz nevus. These lesions have many histologic features of melanoma, although, in children, these nevi classically follow a benign course. Unfortunately, however, unusual lesions with Spitzoid features have been described that have behaved as melanomas. Atypical lesions have been recently referred to as atypical Spitz tumors, some of which represent true Spitzoid melanomas. These tumors are an example of the melanomas that belong to the “unclassified” subtype or, if clinically appropriate, the melanoma of childhood.

Melanoma Staging

The AJCC staging rests upon the TNM classification system to define stage 0 through IV disease. The tumor thickness cutoffs for T1, T2, T3, and T4 disease are defined as in situ disease, with a Breslow depth less than or equal to 1 mm, between 1 and 2 mm, between 2 and 4 mm, and greater than 4 mm, respectively. The “a” and “b” designations are defined as the presence or absence of ulceration for any T or mitoses greater than 1 in T1 disease. The nodal status, “N,” is defined as 0, 1, 2–3, and 4 or more nodes for N0 through N4 disease, respectively. Finally, metastases, M, are defined as M0, M1a, M1b, and M1c based on the absence of metastases; distant skin, subcutaneous, or nodal metastases; lung; or other visceral or distant metastases. The staging system is thus defined as stage I or II if there is absence of nodal involvement, stage III disease if there is nodal involvement in the absence of metastases, and stage IV disease if metastases are present.

Stage I and II nonulcerated melanoma carries a 10-year survival of 95%. With a single mitosis per square millimeter, however, survival decreases to 88%. The inclusion of the mitotic index in the current AJCC is a new addition, and it replaces the Clark level, a histologic description of invasion based on anatomic structures, as an important indicator of survival and lymph node status. Patients with stage III nodal disease demonstrate great variability in outcome based on the tumor burden. Micrometastases in one node carries approximately a 56% 5-year survival rate, while microscopic disease in greater than four nodes carries an approximately 34% survival rate. Stage IV disease, defined as disease-displaying distant metastasis, carries a 5-year survival rate of less than 10%. For this group, chemotherapy and emerging therapies may provide increased survival benefits.