Original Article Diagnostic Utility of P63 and CD10 in Distinguishing Cutaneous Spindle Cell/Sarcomatoid Squamous Cell Carcinomas and Atypical Fibroxanthomas
Jordan M. Hall, Jeff S. Saenger and Oluwole Fadare
Department of Pathology and Laboratory Services, Brooke Army Medical Center, Fort Sam Houston, TX 78235, U.S.A; Department of Pathology, Wilford Hall Medical Center, Lackland AFB, TX 78236, U.S.A.; Department of Pathology, University of Texas Health Science Center at San Antonio, San Antonio, TX 78229
Received 22 Feb 2008; Accepted and available online 7 March 2008
Abstract: The pathologic distinction of atypical fibroxanthomas (AFXs) from cutaneous spindle cell/sarcomatoid squamous cell carcinomas (SCSCCs) may occasionally pose a significant diagnostic challenge, given the substantial clinicopathologic overlap between these lesions. Recent studies indicate that p63 and CD10 are expressed in significant proportions of SCSCC and AFX, respectively. The purpose of this study is to investigate the utility of CD10 and p63 in distinguishing cutaneous SCSCCs and AFXs. The immunohistochemical expression of p63, CD10, cytokeratin AE-1/3, cytokeratin 5/6 and a cytokeratin cocktail (Kermix) was evaluated in an archived group of 23 AFXs and 10 SCSCCs. CD10 was positive in 18/23 AFXs (78%), with most demonstrating strong and/or diffuse staining. Three of 23 AFXs (13%), all negative for cytokeratins, showed focal and weak nuclear staining for p63. Two of 23 AFXs (9%) demonstrated very focal or weak staining for only one cytokeratin; in both cases, p63 and CD10 were negative. One AFX was negative with all immunostains. CD10 was positive in 6/10 SCSCCs (60%), with half demonstrating strong and/or diffuse staining. P63 was positive in 9/10 SCSCCs (90%), with most demonstrating strong and diffuse staining. One SCSCC was negative for p63, but positive with two cytokeratin immunostains. In conclusion, the expression of any of the cytokeratins evaluated herein significantly distinguished AFX from SCSCC. CD10 used in isolation, however, was not useful in making this distinction (positive in 18/23 AFXs versus 6/10 SCSCCs, p=0.4). The addition of CD10 to a panel that includes p63 did not provide any additional information to that obtained from the latter alone. Overall, the most effective combination to distinguish AFX from SCSCC was p63 and cytokeratin AE-1/3. Positivity for both p63 and cytokeratin AE-1/3 was seen in 9/10 SCSCCs (90%) and was not observed in any of the 23 AFXs (p<0.0001). The usefulness of CD10 in this differential diagnosis is limited. (IJCEP802001).
Address all correspondence to: CPT Jordan Hall, MD, Department of Pathology, Wilford Hall Medical Center, 2200 Bergquist Drive, Ste 1, Lackland AFB, TX, 78236, U.S.A, Telephone: (210) 292-7741, Fax: (210) 292-2269, E-mail: Jordan.Hall@lackland.af.mil