Letter to the Editor
To the Editor:
The correspondence titled “Sebaceous carcinoma in-situ”1 touches upon an exceedingly important issue in dermatopathology, that is, the differentiation of carcinoma in-situ from carcinomas that are no longer in-situ. That correspondence1 utilizes as a springboard for discussion both an article that presented a handful of neoplasms with sebaceous differentiation showing a discrepancy in criteria for determining malignancy/benignancy as assessed at scanning magnification from those discerned at high-power magnification2 and my published comments as reviewer of that work.3 Readers of this journal now have three different interpretations of the same collection of cases to contemplate(!): first, the original authors' assessment that the neoplasms with sebaceous differentiation were not further classifiable as benign or malignant,2 second, my interpretation as reviewer of the original work that four were sebaceous carcinoma and one sebaceoma,3 and third, the conclusion that most, if not each of the five cases, are sebaceous carcinoma in-situ rather than sebaceous carcinoma.1 Irrespective of what this implies about the lack of uniform criteria for coming to a diagnosis for certain lesions in dermatopathology, the differentiation between carcinoma in-situ and carcinoma is not merely a semantic distinction for the following reason: carcinomas in-situ lack the capability to metastasize once they have been completely removed, whereas carcinomas that are not in-situ possess that potential. This difference has far-reaching implications for patients and clinicians who manage them.
How does a histopathologist make the determination that a carcinoma is in-situ? Drs Kramer and Chen utilize the features of “well circumscribed with smooth borders” as an indication that the carcinoma is “still confined in epithelium,” that is, it is an in-situ carcinoma. Judging from the size of the neoplasms with sebaceous differentiation that are being discussed, Drs. Kramer and Chen accept lesions that are strikingly larger, and even scores larger, than the original structures they purportedly have replaced (Fig. 1 reprinted from original article).1 In coming to a diagnosis of sebaceous carcinoma in-situ, Drs. Kramer and Chen use analogous findings to those for in-situ carcinomas in breast pathology. I apply different criteria for determining whether a cutaneous neoplasm is still in-situ. In my view, if a carcinoma alters the original cutaneous epithelial structure so that it can no longer be recognized as the original structure, it is no longer in-situ. The neoplasms being discussed (Fig. 1) show no evidence of pre-existing sebaceous lobules or follicular epithelium, therefore, I render a diagnosis of carcinoma rather than carcinoma in-situ. I fully accept that carcinomas in-situ may increase the dimensions of the pre-existing epithelial structure they occupy; however, in these neoplasms with sebaceous differentiation I see no evidence of such structures. I do not dispute the notion of sebaceous carcinoma in-situ; I just disagree with that assessment for these specific neoplasms. For me, the presence of sebaceous ducts and/or individual sebocytes is not confirmation of arising within pre-existing epithelium but instead findings indicative of sebaceous differentiation. I also do not use circumscription or smooth borders as distinguishing criteria between carcinoma in-situ and carcinoma because both of those features may also be seen in carcinomas that are clearly not in-situ, for example, some nodular pattern basal-cell carcinomas or squamous-cell carcinomas that extend into the deep dermis or subcutaneous fat. In no way am I implying that carcinomas in-situ do not extend into deep dermis or subcutaneous fat. Instead, I fully recognize that phenomenon and alert clinicians to it with a comment in the pathology report along the lines of “the lesion extends to the base of the sections/into subcutaneous fat via involvement of adnexal epithelium” because the depth of in-situ involvement has clinical relevance.
In sum, the differentiation between carcinomas that are in-situ and those that are no longer in-situ is more than an intellectual debate because of the different clinical implications of these diagnoses. The criteria I set forth for distinguishing between carcinoma in-situ and carcinoma may or may not be universally employed by dermatopathologists, but I have found that in daily practice they are easily applicable and serve clinicians and patients well.
Kenneth S. Resnik, MD
Institute for Dermatopathology, Newtown Square, PA
1. Kramer JM, Chen S. Sebaceous carcinoma in-situ. Am J Dermatopathol
2. Kazakov DV, Kutzner H, Spagnolo DV, et al. Discordant architectural and cytological features in cutaneous sebaceous neoplasms-a classification dilemma: report of 5 cases. Am J Dermatopathol
3. Resnik KS. Classifying neoplasms with sebaceous differentiation-a reviewer's comments. Am J Dermatopathol