Fine needle aspiration cytology from this mass yielded highly cellular smears with presence of predominantly dispersed cell population and few cohesive clusters of round to polygonal to plasmacytoid cells. These cells had moderate amount of pale basophilic cytoplasm, round to oval monotonous-looking nuclei with inconspicuous nucleoli. Few cells showed prominent nucleoli and some mitotic figures were also seen (Figure 1). A cell block prepared from the aspirated material showed polygonal to round cells with a moderate amount of pink cytoplasm and a single nucleus with prominent nucleoli in some cells. Considering the site and morphology, a diagnosis of malignant myoepithelioma was considered and an urgent excision was advised.

Click Here to Zoom

Figure 1: Giemsa stained cytology smears showing monotonous looking dispersed cell population with some cells with eccentric nuclei (x400).

Meanwhile CT scan of the head and neck was done which revealed that the mass extended to the tail of the parotid while the submandibular salivary gland was unremarkable. The patient was taken up for surgery and intraoperatively the mass was found to be infiltrating the muscle and reaching the tail of the parotid. However, the submandibular salivary gland was not related to the mass. The mass was excised and sent for histopathological examination. It measured 6.5×5×4 cm with a bosselated outer surface and the cut surface showed large areas of haemorrhage and necrosis (Figure 2). H&E stained sections from the tumor showed tumor cells arranged in the form of nests and lobules (Figure 3). Cell morphology was the same as seen in the cell block (Figure 4). Few areas revealed brown- to blackcolored pigment that stained positive for Masson's Fontana, thus confirming that this pigment was melanin. Extensive sampling showed intracytoplasmic pigment in many cells. In view of these findings, a differential diagnosis of malignant melanoma and myoepithelial carcinoma with melanocytic differentiation was considered and immunohistochemistry was used to differentiate the two. The tumor was strongly positive for S100 and HMB 45. It was negative for Calponin, Smooth muscle actin and Cytokeratin thus ruling out myoepithelial carcinoma. More sections were taken to look for presence of salivary gland or lymph node. No lymphoid tissue could be delineated and the adherent salivary gland was unremarkable (Figure 4). A detailed history was taken and complete clinical examination of the patient was carried out including all mucosal sites, skin and ocular examination to look for a possible primary site. The complete systemic examination was unremarkable, thus giving a final diagnosis of malignant melanoma metastatic to subcutaneous tissue with unknown primary site.

Click Here to Zoom

Figure 2: Mass measuring 6.5x5x4 cm with large areas of haemorrhage and necrosis.

Click Here to Zoom

Figure 3: H&E stained section showing low magnification (x100).

Click Here to Zoom

Figure 4: H&E stained section showing prominent nucleoli and occasional binucleated cells (x400).

This problem becomes more complex when it presents at unusual sites with unknown primary sites (MUP). Lymph nodes have been found to be the most common site of MUP in most studies. Axillary lymph nodes are the most common followed by inguinal and cervical nodes⁴. Schlagenhauff et al found cutaneous/subcutaneous metastases to be more frequent, accounting for more than half of all MUPs in a series including 3258 melanoma patients². The prevalence of solitary lesions of melanoma confined to dermal or subcutaneous tissue has been reported to be 0.51%, 0.63% and 0.92% in three large series^2,4,5. The sites included the head and neck, upper extremity, lower extremity, and trunk.

These tumors were extensively investigated for the first time by Das Gupta et al in a series of 47 patients. The criteria proposed by this group for the diagnosis of melanoma of unknown primary have been universally applied till now. According to these criteria, history of previous orbital exenteration and presence of a mole, birth mark, freckle, chronic paronychia, or other skin lesions previously excised, cauterized or desiccated has to be excluded for a diagnosis of MUP. The authors also proposed to examine patients thoroughly including ophthalmoscopy and examination of the anogenital region¹.

Several studies have reported worse prognosis for patients with MUPs^10, while others have reported better^2,6,11 or similar survival rates^3,7. With regard to isolated subcutaneous masses, Schlagenhauff et al reported a 5 year survival of 83%². Anbari et al described 3 patients with isolated subcutaneous nodules, all 3 of which were alive after 8 years^4, Bowen et al reported an 8-year survival rate of 83%⁵ which is higher than expected for Stage III (13-50%) or Stage IV disease (5-17.9%). Our patient is well after 15 months of follow-up.

Several etiologies for MUP have been proposed including 1) an antecedent unrecognized, spontaneously regressed melanoma¹² 2) a previously excised clinically and/or histologically misdiagnosed melanoma, 3) a concurrent clinically unrecognized melanoma, and 4) de novo malignant transformation of an errant melanocyte at a lymph node or visceral site or melanocytes associated with deeper appendageal structures¹.

In conclusion, malignant melanoma is a difficult diagnosis to make particularly when the primary site is not known. Since it is an aggressive tumor, the importance of timely diagnosis is unquestionable. A high index of suspicion, extensive sampling to look for melanin pigment and appropriate immunohistochemistry are the keys to accurate diagnosis.

1) Dasgupta T, Bowden L, Berg JW: Malignant melanoma of unknown primary origin. Surg Gynecol Obstet 1963, 117: 341-345 [ PubMed ]

2) Schlagenhauff B, Stroebel W, Ellwanger U, Meier F, Zimmermann C, Breuninger H, Rassner G, Garbe C: Metastatic melanoma of unknown primary origin shows prognostic similarities to regional metastatic melanoma: recommendations for initial staging examinations. Cancer 1997, 80:60-65 [ PubMed ]

3) Giuliano AE, Moseley HS, Morton DL: Clinical aspects of unknown primary melanoma. Ann Surg 1980, 191:98-104 [ PubMed ]

4) Anbari KK, Schuchter LM, Bucky LP, Mick R, Synnestvedt M, Guerry D 4th, Hamilton R, Halpern AC: Melanoma of unknown primary site: presentation, treatment, and prognosis--a single institution study. University of Pennsylvania Pigmented Lesion Study Group. Cancer 1997, 79:1816-1821 [ PubMed ]

5) Bowen GM, Chang AE, Lowe L, Hamilton T, Patel R, Johnson TM: Solitary melanoma confined to the dermal and/or subcutaneous tissue: evidence for revisiting the staging classification. Arch Dermatol 2000, 136:1397-1399 [ PubMed ]

6) Chang P, Knapper WH: Metastatic melanoma of unknown primary. Cancer 1982, 49:1106-1111 [ PubMed ]

7) Reintgen DS, McCarty KS, Woodard B, Cox E, Seigler HF: Metastatic malignant melanoma with an unknown primary. Surg Gynecol Obstet 1983, 156: 335-340 [ PubMed ]

8) Zelger BG, Steiner H, Wambacher B, Zelger B: Malignant melanomas simulating various types of soft tissue tumors. Dermatol Surg 1997, 23: 1047-1054 [ PubMed ]

9) Lodding P, Kindblom LG, Angervall L: Metastases of malignant melanoma simulating soft tissue sarcoma. A clinico-pathological, light- and electron microscopic and immunohistochemical study of 21 cases. Virchows Arch A Pathol Anat Histopathol 1990, 417:377-388 [ PubMed ]

10) Pack GT, Gerber DM, Sharnagel IM: End results in the treatment of malignant melanoma. A report of 1190 cases. Ann Surg 1952, 136: 905-911 [ PubMed ]

11) Cormier JN, Xing Y, Feng L, Huang X, Davidson L, Gershenwald JE, Lee JE, Mansfield PF, Ross MI: Metastatic melanoma to lymph nodes in patients with unknown primary sites. Cancer 2006, 106:2012-2020 [ PubMed ]

12) Smith JL Jr, Stehlin JS Jr: Spontaneous regression of primary malignant melanomas with regional metastases. Cancer 1965, 18: 1399-1415 [ PubMed ]