Distinguishing Pseudoepitheliomatous Hyperplasia From Squamous Cell Carcinoma in Mucosal Biopsy Specimens From the Head and Neck

Abstract

To the Editor.—I read with great interest the article titled “Distinguishing Pseudoepitheliomatous Hyperplasia From Squamous Cell Carcinoma in Mucosal Biopsy Specimens From the Head and Neck” by Zarovnaya and Black.1 As a dermatopathologist, I frequently encounter a similar issue, especially in evaluation of resection of cutaneous squamous cell carcinoma (SCC) in which a recent prior biopsy has already been obtained. Although I agree with the premise of the authors—that distinguishing pseudoepitheliomatous hyperplasia (PEH) from invasive SCC in small and limited mucosal biopsies is difficult—I do not think that their methods of study warrant the conclusion that certain immunohistochemical stains, namely p53, matrix metalloproteinase 1, and E-cadherin, are useful as a diagnostic adjunct in difficult cases.Their study was based on a retrospective search of head and neck tissues that had previously been diagnosed as either PEH or SCC. It is not clear in their “Materials and Methods,” but it seems that these initial diagnoses were retained in assigning the cases to either group, without reassessment as to whether any of the studied 33 cases were initially misdiagnosed. If indeed the 16 “SCC” and 17 “PEH” cases were correctly diagnosed initially, then the premise that distinguishing SCC and PEH using conventional microscopy is difficult is not supported. In fact, examination of hematoxylin-eosin–stained photomicrographs of the specimens given, for example, in Figures 1, a and 2, a would lead to the correct conclusion, without much difficulty, that Figure 1 represents PEH and Figure 2 represents SCC.However, if some of the 33 cases examined had been reclassified on repeat examination, then the strength of the data presented using the original classification would be diminished, and it is unclear whether the P values presented for the difference in staining for p53, matrix metalloproteinase 1, and E-cadherin between PEH and SCC would remain as low as presented in Table 2. Examination of Figures 3, a and 4, a may lend some support to this possibility because the cytologic atypia seen in Figure 4, a (PEH) appears greater than that in Figure 3, a (PEH with superimposed “mild dysplasia”), whereas logic would dictate that the converse should hold true.More useful would be a similar study using the same stains in the examination of histologically difficult cases of PEH/SCC in which either SCC was diagnosed in the original biopsy and a subsequent specimen showed more definitive evidence of PEH, or the converse, in which PEH was originally diagnosed and a subsequent specimen showed SCC. Such a study would reflect the actual situation in which these stains may or may not have been useful in distinguishing between the 2 entities.