Atypical Fibroxanthoma: Immunophenotypic Lineage Determination and Diagnostic Perspective
Michelle K Horton, Suash Sharma, Wendy B Bollag, Daniel J Sheehan. Georgia Health Sciences University, Augusta, GA
Background: Atypical fibroxanthoma (AFX), typically a plaque/nodule on sun damaged skin, is considered a diagnosis of exclusion, yet requires strict histologic diagnostic criteria. Although it has a histologically aggressive appearance, it remains usually localized to the dermis, with limited extension into subcutis, and is usually cured by complete excision. Currently, there are no reliable immunohistochemical markers to confirm the diagnosis of AFX. Our goal was to determine the utility of smooth muscle actin (SMA), CD271 (p75 NGFR) (a mesenchymal stem cell marker) and SOX-2 in the diagnosis of AFX.
Design: Eleven cases with the diagnosis of atypical fibroxanthomas were collected from our institution's pathology archives between the years of 2006 and 2011. Immunohistochemistry for smooth muscle actin, CD68, SOX-2 and CD271 (p75 NGFR) was performed on formalin-fixed paraffin-embedded archival tissue blocks.
Results: The histopathologic diagnosis was rendered by or verified in consultation with a certified dermatopathologist, and immunohistochemical slides stained and interpreted for SMA, CD271 (p75 NGFR) and SOX-2. Six (55%) of the cases of AFX showed immunopositivity for smooth muscle actin in a varying proportion of tumor cells, including pleomorphic cells, consistent with myofibroblastic/myogenic immunophenotype, given the absence of any hemangiopericytomatous pattern. All eleven (100%) cases expressed CD68 in a significant subset of tumor cells. CD271 showed cytoplasmic with/without membranous staining of any neoplastic cells in 8/11 (72%) cases, of which a significant proportion of cells was positive in 3/11 (27%). However, SOX-2 was uniformly negative in all cases with appropriate internal control. Additional immunostains found to be positive at the time of diagnosis included CD10 (5/5), vimentin (2/2), and factor VIIIA (1/2). Stains that were negative included pan-keratin (9/9), pan-melanoma (5/5), melan-A (2/2), S-100 (9/9), CD34 (3/3), high molecular weight keratin (1/1), CK5/6 (2/2), lysozyme (1/1), CD45 (1/1), and CD30 (1/1).
Conclusions: About half of AFX cases exhibit focal or diffuse myofibroblastic/myogenic immunophenotype, which in the context of CD271 positivity in at least a subset of tumor cells, in the absence of SOX-2 expression, suggests derivation of this neoplasm from mesenchyme-committed stem cells with secondary myofibroblastic/myogenic differentiation.
Tuesday, March 20, 2012 9:30 AM
Poster Session III # 88, Tuesday Morning