TTF-1, Napsin A, and Surfactant A to Identify Lung Carcinoma in Malignant Pleural Effusions – Are More Stains Better Than One?
MR Phillips, AE Walts. Cedars-Sinai Medical Center, Los Angeles, CA
Background: Lung carcinoma (LC) is the leading cause of cancer deaths in the U.S. The lungs are also common sites of metastases from extrapulmonary primaries (EXP). Optimal clinical management often requires clarification of the origin of tumor cells in malignant pleural effusions (PE). TTF-1 (TTF1), napsin A (napsin), and surfactant A (surfactant) immunostains are used to support a diagnosis of LC. TTF1 is frequently the initial stain ordered given that thyroid carcinoma rarely involves PE. This study evaluates napsin and surfactant expression in a series of PE with TTF1 positive (TTF1+) and TTF1 negative (TTF1-) LC and EXP.
Design: Serial sections of formalin fixed, paraffin embedded cell blocks from 92 PE with histologically and/or clinically confirmed LC (N=58) and EXP (N=34) were immunostained for TTF1, napsin, and surfactant. LC PE included 30 TTF1+ and 28 TTF1- cases, and EXP PE included 30 carcinomas (10 breast, 5 mullerian, 3 pancreas, 3 colon, and 1-2 each of stomach, prostate, cervix, larynx, kidney, bile duct) and 4 mesotheliomas. Staining intensity (weak (1+), moderate (2+), strong (3+)) and percentage of tumor cells stained were recorded for each case. Staining was considered positive when intensity was 1+ in ≥10% of tumor cells or ≥2+ in any number of tumor cells. Results were correlated with final diagnoses.
Results: Among the TTF1- LC PE, 8 (29%) and 12 (43%) cases were positive for napsin and surfactant, respectively. Both stains were positive in 7 of these cases. Among the TTF1+ LC PE, all 3 stains were positive in 20 (67%) cases, with napsin positive in 24 (80%) and surfactant positive in 23 (77%) cases. More than 75% of tumor cells were positive in 80% (24/30) of all TTF1+, 84% (27/32) of all napsin positive, and 35% (14/40) of all surfactant positive malignant PE. Among the EXP PE, TTF1, napsin and surfactant were negative in 34 (100%), 34 (100%), and 29 (85%) cases, respectively. Five (15%) of the EXP PE (2 colon, 2 breast, 1 cervix) were positive for surfactant, with staining observed in 5%-60% of tumor cells.
Conclusions: TTF1 and napsin exhibited similar sensitivities and specificities indicating that they are comparable as initial stains for LC in PE. Both napsin and surfactant identified several of the TTF1- LC PE; however, the 100% specificity of napsin rendered it more reliable than surfactant as an initial and 2nd tier stain for LC. When both TTF1 and napsin are negative, a positive stain for surfactant should only be interpreted in the context of a broader panel of immunostains.
Wednesday, March 24, 2010 1:00 PM
Poster Session VI # 61, Wednesday Afternoon