[927] Establishment of Laminin and Pankeratin Dual Immunostain for the Evaluation of Urothelial Carcinoma

DL Zynger, OM Radu, AV Parwani. University of Pittsburgh Medical Center, Pittsburgh, PA

Background: Incorrect staging due to misinterpretation of lamina propria invasion is one of the most commonly described diagnostic errors in bladder specimens. Lamina propria invasion can be difficult to assess due to the presence of only a small microinvasive focus or fragmentation artifact. Immunohistochemical assessment for basement membrane or epithelial cells may help corroborate the histologic impression of lamina propria involvement. We sought to create a dual immunostain containing a basement membrane marker (laminin) and an epithelial marker (pankeratin) to facilitate the interpretation of lamina propria invasion.
Design: 504 cases of paraffin embedded urothelial carcinoma were obtained, including 58 whole tissue sections (cystectomies, transurethral resections and biopsies) and 446 cases using tissue microarray. The tissue was incubated with laminin primary monoclonal antibody (1:25) with 3,3'-diaminobenzidine as the chromogen followed by pankeratin cocktail incubation (prediluted) using Vulcan fast red as the secondary chromogen. Pankeratin was semi-quantitatively analyzed as 0, < 5% of cells stained; 1+, 5-10% of cells stained; 2+, 11-50% of cells stained; or 3+, >50% of cells stained. Invasive and noninvasive urothelial components as well as blood vessels were assessed for laminin.
Results: Pankeratin stained benign and malignant urothelial cells in the vast majority of the 504 cases (0, 2%; 1+, 1; 2+, 1%; 3+, 96%) with intense staining and little background, allowing for clear visualization. Within the 58 whole tissue sections, 40 contained only invasive carcinoma, all of which completely lacked laminin expression surrounding tumor. 18 of the 58 tissue sections contained a noninvasive component, demonstrating laminin separating the urothelium from the lamina propria. Dual staining highlighted vascular invasion in 5 of the 58 cases. One of the 5 cases containing vascular invasion was not diagnosed in the corresponding pathology report.
Conclusions: We created a dual immunostain of laminin and pankeratin. Invasive carcinoma was strongly positive for pankeratin and lacked laminin. Noninvasive components showed dual expression. The dual immunostain of laminin and pankeratin shows promise as a tool to facilitate the interpretation of invasion in bladder specimens. Additionally, this dual stain highlights vascular invasion, allowing for the assessment of both lamina propria and vascular invasion on the same slide.
Category: Genitourinary (including renal tumors)

Monday, March 9, 2009 1:00 PM

Poster Session II # 122, Monday Afternoon