Lymph node metastases of melanoma.

The incidence of melanoma is increasing rapidly among Caucasians in America, Europe, and Australia. In the US, melanoma is increasing at a rate that exceeds all other neoplasms except for lung cancer in women (1). The skin pigmentation is protective and incidence rates are considerably lower in African-American, Hispanics, and Asians (1). In 2001, the estimates in the US were for 51,800 new cases and 7800 deaths (1). Melanoma is rare in children, median age at diagnosis is 53 years (1).

Sunlight, immunological conditions, and genetic abnormalities are the major factors in the origin of melanomas (2). As they evolve, after variable amounts of time, some malignant tumors acquire the ability to disseminate and form new tumors at distant sites. Clark (3), who has studied melanocytic neoplasia extensively, have offered concepts of tumor progression derived from this system that may be applicable to all malignant neoplasia. Melanocytic lesions are pigmented and superficial, and therefore readily apparent and easy to follow. According to his concepts, melanomas arise by generic steps of tumor progression, from melanocytic hyperplasia in the acquired nevus, to aberrant differentiation marked by abnormal hyperplasia and nuclear atypia in the dysplastic nevus, to a primary malignant neoplasm that lacks the ability to metastasize in the radial growth phase, and eventually to a malignant neoplasm endowed with the capacity to metastasize in the vertical growth phase. Morphologically, this crucial transformation occurs when radial growth changes to vertical growth. Thus, for a primary neoplasm to become metastatic, a critical and qualitative change must take place, in which a new population of cells with new properties originates within the neoplasm (3). In most instances, the melanocytes of an acquired nevus follow a pathway of differentiation that leads to their disappearance (3). The transformation from non-neoplastic to neoplastic, from noninvasive to invasive, and eventually from nonmetastatic to metastatic occurs in the melanoma system, as in most other cancers, by processes of mutation and clonal selection.

In the skin, each melanocyte contacts through dendritic processes about 20 to 30 keratinocytes (2). As in other cancers progression to invasion is characterized by changes in cell-cell and cell-matrix adhesion and communication. A main feature in this process is the loss of E-cadherin expression and increase in N-cadherin which coincides with the escape of melanocytes from keratinocyte control and further progress to invasion and metastasis (4).

The studies of Breslow have shown that when the thickness of a cutaneous melanoma exceeds 0.76 mm, the risk for the development of metastases increases in proportion to tumor thickness (5). In Breslow’s series of 138 patients, 39% had primary lesions with a thickness of less than 0.76 mm, and all survived free of disease for 5 years or more, whereas metastases developed in 33% to 84% of those with lesions thicker than 0.76 mm.

The prognostic criteria based on these observations have been confirmed repeatedly. In a large study of prognostic factors in cutaneous melanoma, 90% of patients with superficially invasive, thin primary tumors were alive without recurrence 5 years after surgical excision of the skin lesion (6). In a minority of patients, metastases develop in the regional lymph nodes and, when they do, their chances of survival drop to only 40% at 5 years after lymphadenectomy. In almost all cases, when a cutaneous melanoma reaches the phase of dissemination, it spreads to the regional lymph nodes before it spreads to more distant sites (9,10). Therefore, the progression from stage I to stage II is a very important step in the evaluation of melanoma that must be documented by careful investigation of the regional lymph nodes. A study of 2,227 lymph nodes from 100 patients with clinical stage I cutaneous melanoma revealed no metastases on sections stained with hematoxylin and eosin, yet melanoma cells were detected in 16 nodes from 14 patients when immunostaining with antibody to S100 protein was used (8). A possible pitfall in applying this technique to identify metastatic melanoma cells is that they can be confused with nodal interdigitating reticulum cells, which also express S100 protein (8,11).

Patients with stage II melanoma present with a painless mass that in one study ranged in size from 1 to 18 cm in diameter (median, 4 cm) (20) (Fig. 87.1). A particular aspect of melanomas that causes major difficulties in diagnosis and prognosis is occasional spontaneous regression. This phenomenon, probably related to complex interrelations with tissue immunity, is more common in melanomas than in any other malignant tumor. As a result, some cutaneous melanomas with a thickness of less than 0.76 mm are associated with lymph node metastases (21). Even more difficult to interpret are metastatic melanomas with occult primary tumors, estimated to represent about 4% of cases (22,23). In general, metastatic cancers with an unknown primary site are not uncommon and may pose an extremely frustrating problem. They were ranked as the eighth most common site of cancer in a large series of cases accounting for 10% to 15% of referred patients with solid tumors (24,25). In a study of 1,539 patients with an unknown primary cancer, melanomas represented 2.9% of cases (26). The male-to-female ratio was 1.6:1.0 and the median age was 48 years (no patient was younger than 15 years) in a series of 166 cases of lymph node metastases with an unknown primary (20). In descending order of frequency, the lymph node groups involved by metastatic melanoma were axillary (47%), inguinal (20%), and cervical (24%). In women, inguinal metastases were almost as common as axillary metastases (15). The 5- and 10-year survival rates of patients with lymph node metastatic melanoma were 46% and 41%, respectively. These were not influenced by age of the patient, size of the metastases, or pigmentation (melanotic vs. amelanotic). However, patients with axillary metastases had consistently better survival rates than those with metastases in other locations (20). Patients with metastases to a single lymph node had significantly better 5-year survival rates than those with more than one involved lymph node (21). In a retrospective study of 3,805 patients
with melanomas at the Memorial Sloan-Kettering Cancer Center, 166 had no known primary tumor, a prevalence of 4.4% (20). The 5-year survival rates of patients who had metastatic melanoma in one group of regional lymph nodes and were treated by lymphadenectomy were similar for those with a known and those with an unknown primary (22). This finding, subsequently confirmed by others (23), suggests that the occult primaries must have been melanomas that regressed, precluding the occurrence of additional metastases. A rare exception is a primary melanoma in a lymph node arising from a prior lymph node nevus, documented in a case report (28).