Myoepithelial cell cocktail (p63+SMA) for the evaluation of sclerosing breast lesions

Abstract

Sclerosing breast lesions may sometimes mimic the appearance of infiltrating carcinoma due to the entrapment of ductular structures in a fibrotic core. The immunohistochemical detection of the outer myoepithelial cell layer that is indicative of a non-infiltrating lesion is a valuable clue for the diagnosis of such ambiguous cases. The myoepithelial cell markers smooth muscle actin (SMA) and p63 are most commonly used since their specificity and sensitivity are well established. However, recent studies have indicated that some morphologically distinct myoepithelial cells fail to stain for SMA and that p63 positivity can be rarely expressed by a subset of malignant epithelial cells. Moreover, SMA can also be positive in stromal myofibroblastic cells and normal vessels that can be found close to the entrapped ductules and might be erroneously interpreted as myoepithelial cells. In this study, we used a double-immunolabeling technique combining both SMA and p63 antibodies (myoepithelial cell cocktail), in order to investigate whether this technique is advantageous over either marker used alone, in diagnosing sclerosing breast lesions. Our results indicate that p63 alone is not a useful myoepithelial cell marker if applied in large sclerosing breast lesions, however, in smaller lesions it is still of high value. On the contrary, SMA proved significantly useful in the evaluation of myoepithelial cells in larger but not in smaller complex sclerosing lesions. The myoepithelial cell cocktail has a staining sensitivity identical to that of SMA. Nevertheless, in a certain number of cases the cocktail might be useful in differentiating myoepithelial cells from stromal myofibroblasts or vascular smooth muscle cells due to the false impression of a higher staining intensity of the cocktail resulting from the expression of both nuclear and cytoplasmic/membranous antibodies that occupy a wider area of the cell under control.