Original ArticlesExpanding the Morphologic Spectrum of Differentiated VIN (dVIN) Through Detailed Mapping of Cases With p53 LossSingh, Naveena MB, BS, FRCPath*; Leen, Sarah L. MB, BS, FRCPath*; Han, Guangming MD, FRCPC†; Faruqi, Asma MB, BS, FRCPath*; Kokka, Fani MB, BS, MRCOG*; Rosenthal, Adam MB, BS, MRCOG, PhD‡; Jiang, Xin Rong§; Kim, Rachel MD∥; McAlpine, Jessica N. MD, FRCSC∥; Gilks, C. Blake MD, FRCPC§Author Information Departments of *Cellular Pathology ‡Gynecological Oncology, Barts Health NHS Trust, London, United Kingdom †Department of Pathology, Sunnybrook Health Sciences Centre, Toronto, ON §Department of Pathology ∥Division of Gynecological Oncology, Vancouver General Hospital, Vancouver, BC, Canada Supported by an unrestricted educational grant from Sanofi Canada. The authors have disclosed that they have no significant relationships with, or financial interest in, any commercial companies pertaining to this article. Correspondence: C. Blake Gilks, MD, FRCPC, Room 1259, 1st Floor JPPN, Department of Pathology, Vancouver General Hospital, 855 West 12th Ave, Vancouver, BC, V5Z 4E3 Canada (e-mail: firstname.lastname@example.org). The American Journal of Surgical Pathology: January 2015 - Volume 39 - Issue 1 - p 52-60 doi: 10.1097/PAS.0000000000000291 Buy Metrics Abstract The pathogenesis of vulvar squamous cell carcinoma follows 1 of 2 distinct pathways. A precursor lesion in the human papilloma virus–independent pathway, differentiated vulvar intraepithelial neoplasia (dVIN), was only recently characterized in detail and is infrequently diagnosed without an associated component of invasive carcinoma. Aberrant p53 immunostaining is frequently seen in dVIN, and in approximately 25% to 30% of cases it manifests as a complete loss or a p53-null pattern. The abrupt transition between p53 loss and basal p53 expression in lesional versus nonlesional epithelium allows clear demarcation between neoplastic and non-neoplastic epithelium. For this study, 14 specimens from 10 patients were identified from the pathology archives of 2 teaching hospitals on the basis of: (1) a diagnosis of dVIN, with or without invasive carcinoma; and (2) p53-null immunostaining pattern in lesional cells. Ten specimens had associated invasive carcinoma. All sections from each specimen that showed the specimen resection margin were stained for p53 and reviewed together with all hematoxylin and eosin sections. Detailed morphologic assessment of the p53-null epithelium was made and compared with the adjacent benign squamous epithelium. The status of the resection margins based on the original pathologic assessment was compared with that assessed with p53 immunohistochemistry. One specimen showed p53 loss in the invasive carcinoma but patchy basal positivity in the region originally diagnosed as dVIN, supporting interpretation as a benign hyperplastic focus, rather than dVIN. In the remaining 13 specimens the areas originally diagnosed as dVIN, as well as the associated invasive carcinoma (if present), were p53-null. In 8 of these specimens, on the basis of the presence of p53-null immunostaining and subtle morphologic abnormalities, dVIN was more extensive than originally recognized. The spectrum of morphologic changes in p53-null regions that were in continuity with areas originally recognized as dVIN were subtle and typically consisted of an abrupt change in maturation of the squamous epithelium (loss of keratohyaline granules and parakeratosis), tinctorial alterations in the keratinocytes, with cells containing more abundant eosinophilic cytoplasm, and minimal basal nuclear atypia. Margin status changed from negative to positive in 4 of 13 specimens and from focally to more extensively positive in an additional 3 specimens. In summary, the clonal in situ component of non–human papilloma virus vulvar squamous cell carcinoma can be characterized by very subtle morphologic abnormalities that may be misinterpreted as benign change. This results in underestimation of the extent of dVIN, and, as a result, resection margin involvement may be significantly underestimated. dVIN can also be overdiagnosed in areas of reactive change. Better tools for diagnosis of dVIN are needed; until such tools are developed the limitations in the current diagnosis of dVIN should be recognized. © 2015 by Lippincott Williams & Wilkins.