Carcinoma versus cytokeratin-positive lymphoma: a case report emphasizing the diagnostic role of electron microscopy.

Lymphoma diagnosis rarely needs electron microscopy (EM), but one area where it can be useful is in the distinction of cytokeratin-positive lymphoma from carcinoma. The authors describe such a case, where difficulties were encountered due to lack of antibody specificity, distinguishing reactive from tumoral cells, and suboptimal sampling for EM. The tumor was in a lymph node next to the right submandibular gland in a 69-year-old man. This was a malignant tumor, composed of sheets of monomorphic large round cells. Interpretation on the part of a team of pathologists who examined this tumor was divided. On histological sections, the differential diagnosis was between carcinoma and lymphoma, which was modified to cytokeratin-positive lymphoma versus carcinoma since tumor cells were found to be cytokeratin-positive. EM of tumor retrieved from formalin showed plasmablastic features, in keeping with lymphoma with plasmablastic differentiation, one of the light microscopy diagnoses. The moderately strong positivity of cytokeratin and the positivity for Ber-EP4, however, supported carcinoma, and further sampling for EM was carried out, specifically on a cytokeratin-positive area of the wax block. Tonofibrils were found, supporting carcinoma. The final diagnosis was undifferentiated carcinoma with unknown primary site. The study emphasizes the need to take into account the imperfect specificity of cytokeratin, which can be found in several hemolymphoid neoplasms, to distinguish reactive from neoplastic cells, and to secure appropriate sampling for EM. This is one of the occasions where dewaxing (of an immunohistochemically defined wax block) offers positive advantages, despite compromised structural preservation, in the search for diagnostically important organelles.