To describe the clinicopathologic features of melanocytic proliferations associated with lichen sclerosus (LS) and to compare these findings with those in controls.

A 44-year-old female developed confluent, dusky red, pruritic labial papules clinically suspected to be genital warts. She had a long-standing history of Crohn's disease with vulvar fistulae. The papular eruption developed after several bouts of cellulitis in a region of vulvar lymphedema. Shave biopsy of a papule exhibited papillated epidermal hyperplasia overlying a dermis with a 'Swiss-cheese' appearance secondary to lymphedema and superficial ectatic thin-walled vascular spaces characteristic of lymphangiectasias. Review of published cases reveals that acquired lymphangiomas often affect the vulva compared to other cutaneous sites and can be associated with surgery, radiation therapy, infection (e.g., erysipelas, tuberculosis), Crohn's disease, congenital dysplastic angiopathy and congenital lymphedema. Rather than translucent vesicles ('frog spawn') typical of extragenital cutaneous lymphangiomas, vulvar lymphangiomas often present as verrucous papules that can be mistaken for genital warts. In this case, we believe that the combination of vulvar Crohn's disease and recurrent cellulitis resulted in local lymphatic destruction, lymphedema and ultimately symptomatic lymphangiectasias that mimicked genital warts.

Lichen sclerosus (LS) shares with vitiligo a milky-white appearance. By biopsy, pathognomonic dermal sclerosis readily distinguishes LS from vitiligo and other causes of leukoderma. To determine what the mechanism of hypopigmentation is in LS, we examined samples from LS cases for alterations in melanin content (Fontana-Masson stain) and melanocyte number (HMB-45 [PMEL-17/gp100], Mel-5 [TRP-1], Mart-1 [Melan A]) and compared these findings with those in controls of normal skin, acute scars, vitiligo, and lichen planus (LP; a common inflammatory cause of hyperpigmentation). The degree and extent of melanization found in LS overlapped with that in acute scars showing predominantly hypomelanized keratinocytes, with that in LP containing regions with numerous melanophages, and with that in vitiligo exhibiting focal regions of keratinocytes devoid of melanin pigment. By hematoxylin-eosin staining and immunocytochemistry for Mel-5 and Mart-1, LS had a lower mean count of melanocytes than acute scars, LP, and normal skin per 200 basal keratinocytes. In addition, a few LS cases had a significant loss of melanocytes comparable to that of vitiligo. Surprisingly, Mart-1 identified rare melanocytes in 67% of vitiligo cases and a significantly larger pool of melanocytes in LS and controls other than those labeled by Mel-5. Furthermore, LP and evolving lesions of LS contained the highest Mart-1 counts. HMB-45-immunoreactive melanocytes were found in the majority of acute scars and in LP and late-stage LS lesions at significantly lower levels than Mel-5- and Mart-1- labeled melanocytes, but they were not found in vitiligo or normal skin. We propose that several mechanisms may play a role in the production of leucoderma in LS: 1) decreased melanin production; 2) block in transfer of melanosomes to keratinocytes; and 3) melanocyte loss. The latter finding may be the pathogenic connection (lichenoid dermatitis of LS triggering an autoimmune reaction to melanocytes) that underlies the documented association of LS with vitiligo.

Malignant melanoma (MM) is considered to be a chemotherapy-refractory tumor. New anti-cancer drugs (e.g. etoposide) that target DNA topoisomerases (e.g. topoisomerase II-alpha (topo IIalpha)) show activity against a wide variety of solid tumors. In this study, we investigated the frequency and rate of labeling for topo IIalpha in 163 MMs (primary and metastatic) and 67 melanocytic nevi to determine whether topo IIalpha expression is elevated in MM. Primary MM exhibited significantly more frequent topo IIalpha expression compared to benign nevi (86% vs. 56%, p=0.0001). The rate of topo IIalpha labeling in dysplastic melanocytic nevi, radial growth phase MM, vertical growth phase MM and metastatic MM revealed significant differences amongst groups and a positive covariance with advancing stage (means: 0.3, 0.5, 5, and 8 '+' cells/hpf, respectively; r=0.3, all p < or = 0.02). Topo IIalpha labeling significantly correlated with increasing mitotic activity, depth of invasion and Clark's level, diminishing tumor infiltrating lymphocytes, and poor outcome (all p < or = 0.01) in primary MM. For metastatic MM, a minority (30%) exhibited marked elevation of topo IIalpha expression. These findings indicate topo IIalpha as a potential therapeutic target and marker for MM. Immunohistochemical analysis of disseminated MM may allow for correlation with clinical response and enable selection of candidates sensitive for specific chemotherapy.