Tricholemmal carcinoma (TLC) is a cutaneous adnexal tumor with presumed external hair sheath differentiation. In order to better understand the salient features of this neoplasm, we analyzed the histologic and clinical findings in 10 cases of TLC. Eight patients were males, and two were female; they ranged in age from 55-88 years. Each tumor occurred in hair bearing, sun-exposed skin, and involved the scalp, face, trunk, or upper extremities. The lesions were usually slightly raised, pale tan or reddish, and keratotic; were usually present for less than 1 year; and measured 0.4-2.0 cm. All of them were treated by wide local excision; neither recurrence nor metastasis was reported after 11 to 92 months of clinical followup. Histologically, each TLC was composed of a lobular proliferation centered on the pilar apparatus. Cells with glycogen-rich, mucin-negative, clear or pale eosinophilic cytoplasm predominated. Brisk mitotic activity (4-39 mitoses per 10 high power fields) was typical. Involvement of the interfollicular epidermis was invariably noted, with superficial ulceration in seven tumors. Transitional zones between TLC and the adjacent epidermis were not seen, although pagetoid spread occurred in two examples. Invasion of reticular dermis was present in eight cases, with infiltration to mid-dermis in five TLC. All tumors exhibited areas of tricholemmal type keratinization; dyskeratotic cells were noted in six examples. Hyperkeratosis and parakeratosis were variably present as well. Actinic damage was a constant feature. Despite local invasion at diagnosis, the clinical course of TLC was indolent in all cases.

Abnormalities of mucin-type glycoproteins have been described in lung cancers, but their molecular basis is unknown. In this study, mucin-core-peptide-specific antibodies and cDNA probes were used to determine the relative expression of mucin genes corresponding to one membrane-bound mucin (MUC1), two intestinal mucins (MUC2 and MUC3), and one tracheobronchial mucin (MUC4) in normal (nonneoplastic) lung, and in lung neoplasms. Normal lung tissues exhibited a distinct pattern of mucin gene expression, with high levels of MUC1 and MUC4 mRNA and low to absent levels of MUC2 and MUC3 mucin immunoreactivity and mRNA. In contrast, lung adenocarcinomas, especially well-differentiated cancers, exhibited increased MUC1, MUC3, and MUC4 mRNA levels. Lung squamous-cell, adenosquamous, and large-cell carcinomas were characterized by increased levels of MUC4 mucin only. We conclude that the expression of one membrane-bound and several secretory-type mucins is independently regulated and markedly altered in lung neoplasms. The frequent occurrence of increased MUC4 transcripts in a variety of non-small-cell lung cancers indicates the potential importance of this type of mucin in lung cancer biology.

In an attempt to characterize the immunocytochemical attributes of eccrine sweat gland carcinoma, we studied 32 examples of this tumor with antibodies to epithelial membrane antigen (EMA), cytokeratin (CK), carcinoembryonic antigen, S100 protein, alpha-lactalbumin, salivary amylase, blood group isoantigens, beta-2-microglobulin, and Leu M1. All cases expressed EMA and CK, and 28 of 32 cases also displayed at least 2 of the 6 remaining antigens. No significant variations were noted in the immunophenotypes of histologic subtypes of eccrine carcinoma. These results provide an objective means of diagnostic separation between sweat gland carcinoma and other primary malignant cutaneous tumors. However, they do not appear to correlate with the degree of tumoral differentiation, and are of no assistance in the separation of benign and malignant sudoriferous neoplasms. The ability of immunocytochemical techniques to distinguish between primary malignant adnexal cutaneous tumors and metastases to the skin appears unlikely, but remains to be studied further. Also, the use of immunostaining panels is advised in the study of adnexal carcinomas, since no single determinant in isolation is specific for these neoplasms.

No abstract