Immunohistologic differential diagnosis of basal cell carcinoma, squamous cell carcinoma, and trichoepithelioma in small cutaneous biopsy specimens

Abstract

The distinction between squamoid basal cell carcinoma and basaloid squamous cell carcinoma (or between BCC and trichoepithelioma variants) is usually made readily on the basis of defined histological criteria. However, these differential diagnoses occasionally can pose difficult morphological problems. The stated distinctions are clinically important because the risk of progressive disease is significantly higher with squamous carcinoma of the skin than with basal cell carcinoma (BCC), and a trichoepithelioma misinterpreted as BCC burdens the patient with an inaccurate diagnosis that may result in inappropriate surgery. Recent reports have suggested that reactivity with the monoclonal antibody Ber-EP4 is capable of separating histologically similar basal cell and squamous carcinomas, and that the expression of bcl-2 or CD34 antigen is able to distinguish BCC from trichoepithelioma. However, corroborative studies of these contentions are few in number. In order to investigate the usefulness of the stated immunostains in the above-cited differential diagnoses, the authors analyzed 45 basal cell carcinomas and 22 squamous carcinomas, as well as 36 trichoepitheliomas. The monoclonal antibodies Ber-EP4, My10 (CD34), and anti-bcl-2 were applied to formalin-fixed paraffin sections in all cases, using a standard avidin-biotin-peroxidase complex method. Most BCCs demonstrated strong, diffuse cytoplasmic labeling with Ber-EP4 and anti-bcl-2. In contrast, the squamous carcinomas were uniformly negative for the former marker and only focally reactive for the latter in four examples. 'Peripheral' bcl-2 staining of trichoepitheliomas was noted in 24 of 33 of the immunoreactive tumors, but the remainder were marked diffusely and similarly to most BCCs. Among the latter, immature trichoepitheliomas were diffusely reactive for this marker in 6 of 8 cases. Labeling of epithelium for CD34 failed to discriminate between any of the tumor types under evaluation, whereas staining of peritumoral stroma was characteristic of the majority of trichoepitheliomas and more than one-third of metatypical basal cell carcinomas. These data support the suggestion that Ber-EP4 and bcl-2 are useful in the separation of BCC from squamous carcinomas. Nevertheless, they also serve to caution against reliance upon bcl-2 and CD34 immunostains in attempting to distinguish BCC from trichoepithelioma in histologically enigmatic cases. There is currently no certain method other than conventional microscopy that can be applied successfully to the latter problem.